This workshop equips STEM laboratory educators with the tools to design and evaluate effective Intended Learning Outcomes (ILOs) aligned with the principles of Constructive Alignment (CA) and structured through the SOLO taxonomy (Structure of Observed Learning Outcomes). Participants will learn to critically assess and reformulate their laboratory ILOs to reflect progressive levels of student understanding, from surface to deep learning, thus promoting student-centered learning. In a hands-on format, attendees will first (re)develop ILOs for their own (digital) laboratories and then design corresponding pre- and post-tests to systematically assess learning progress. The workshop emphasizes how CA connects ILOs, Teaching and Learning Activities (TLAs), and Assessment Tasks (ATs) to ensure coherence and transparency in laboratory pedagogy. Through a guided example of a cross-reality chemical extraction laboratory, participants will explore how to construct meaningful assessments that capture changes in students’ declarative and functioning understanding. By the end of the session, educators will be able to design laboratory experiences that not only enhance student motivation and autonomy but also provide measurable evidence of learning gains, fostering deeper engagement and reflective teaching practices in STEM laboratory education.

Focusing on mental health, equity and inclusion, cultural diversity, and quality learning outcomes has led to the realisation that respect and supportive university environments are important. Any improvements will not only benefit equity groups but all students, as learning is also about social interaction.

Respect fosters inclusive and supportive learning environments that support students and staff in learning and academic discourses. This is particularly relevant in the current context of polarised public and social media interactions. In universities' quest for knowledge, respect is essential to ensure well-being, diversity of thought and meaningful discourse.

The purpose of this workshop is to develop a shared understanding of what respect looks and feels like in a modern educational context.

Participants will work in small groups to unpack key concepts, such as what respect is. How is respect given and received? And must respect be earned, and can be lost?

It is anticipated that this will provide participants with opportunities to reflect on their own understanding, experiences, and perceptions of respect, and how these influence their interactions with peers and students.

Virtual Reality (VR) is increasingly recognized as a powerful medium for learning, capable of transforming how complex spatial and scientific phenomena are communicated. In higher education, project-based learning and research-integrated teaching approaches enable students to acquire both disciplinary knowledge and transversal skills through authentic design challenges. In parallel, VR is gaining momentum in science communication for public engagement, helping audiences visualize invisible or abstract processes such as geological or energy-system dynamics. While both strands - educational VR and public-facing VR - are well studied independently, little is known about how they can be connected to form a continuous learning structure that benefits multiple audiences.

This paper presents such a connection through a case study of a university VR course linked to public science communication events on geothermal energy. A virtual “elevator into the underground,” developed by students and refined for exhibition use, served as a shared artefact across formal and informal learning contexts. Qualitative evaluation following Kirkpatrick’s model showed high motivation and significant learning gains among students, as well as strong engagement and knowledge transfer among public participants. The results highlight how authentic sustainability contexts and the reuse of student-created artefacts can foster linked learning ecosystems that bridge education, research, and public engagement.

Capture the Flag (CTF) exercises are widely used in cybersecurity education for developing problem-solving and technical skills, but traditional formats often lack realism, teamwork, and situational context. This study, conducted within the ATHENA project at Tallinn University of Technology, explores the use of Virtual Reality (VR) to enhance CTF challenges. Two pilot scenarios were tested: Tower Cipher, a 3D exploration puzzle designed for secondary and vocational students, and Control Room Mission, an industrial control room simulation tested with international cyber talents in both computer-based and VR modes.

A mixed-methods evaluation combined surveys, reflections, and task performance to assess usability, engagement, teamwork, and perceived skill development. Results show that VR tasks significantly improved motivation, collaboration, and situational awareness, especially in operational technology contexts, though technical skill acquisition remained stronger in traditional screen-based CTFs. Challenges included navigation, VR comfort, and scalability.

The findings suggest that VR-enhanced CTFs are not a replacement but a valuable complement to conventional cybersecurity training. They offer promising opportunities for experiential learning, helping bridge the gap between technical exercises and realistic, collaborative practice.

Picking services are essential in logistics and supply chain management. Among these service providers, manual packaging workstations are vital in situations where full automation is not feasible due to economic or operational reasons. This paper presents an overview of A/B testing in VR design projects and affordable ergonomics measurement techniques, which can be used in education as well as in industry. The concept develops and assesses a virtual reality A/B-testing framework that facilitates an iterative, data-driven approach to workstation design, helping to avoid expensive physical prototyping and which allows students to create said scenarios with limited resources available. The modular pipeline includes a virtual prototype stand with standardized task logic, a controlled experimental protocol across six workstation variants, automated collection of head- and hand-trajectory data, automatically computed key performance indicators related to efficiency and ergonomics, and subsequent inferential analyses. Following the implementation, 20 participants empirically test various configurations multiple times to investigate learning effects on times, movement distances, and strain. In addition to the empirical findings, the work provides a reusable infrastructure for VR-based workstation engineering, facilitating rapid evidence-based iteration and a reduction of real-world prototyping.

Virtual reality and metaverse technologies hold strong potential for engineering education, particularly for improving spatial understanding, experiential learning, and collaboration. Yet most implementations remain confined to isolated pilot projects that lack institutional alignment and long-term sustainability. Existing research focuses mainly on single-course case studies, leaving a clear gap in scalable approaches for faculty-wide integration. This paper addresses that gap by presenting a structured, top-down methodology for adopting immersive technologies across an engineering faculty. The proposed Faculty-Level Integration Methodology (FLIM) follows an iterative three-phase process of analysis, design, and validation. The framework is supported by low-cost infrastructure and reusable open-source applications. Results indicate strong motivation among faculty and students, reveal key barriers, and identify opportunities for sustainable institutional adoption, including future university-wide scaling.

Artificial intelligence (AI) is becoming an integral part of daily life, shaping how people access information, consume media, and interact with technology. This study explores how hands-on robotics workshops enhance children's AI literacy in an informal learning environment. A mixed-method approach was used to analyze self-reports on enjoyment and difficulty, along with behavioral observations related to initiative, perseverance, and frustration management. The findings indicate that parental participation influences learning experiences, either by fostering independent problem solving or, in some cases, unintentionally hindering a child's participation. The findings emphasize the need for early AI education that clearly differentiates between automated behaviors and genuine machine learning processes.

Software development projects increasingly require project managers

who can interpret complex data, anticipate risks early, and make informed decisions

in rapidly changing Agile environments. Yet traditional project management

education—both university-based and vocational—often lacks realistic,

data-rich learning contexts where learners can analyze project behavior, explore

risk scenarios, and understand how estimation errors propagate. This paper presents an AI-based dashboard that supports early risk detection in Agile software

projects while simultaneously serving as a training platform in formal project

management (PM) education. The dashboard integrates ML models to predict

task duration deviations, detect sprint-level risks, and provide interpretable analytics.

Beyond real-time predictive insights, the dashboard functions as a learning

simulator for PM training programs. Students can upload or simulate project datasets, test planning assumptions, adjust risk thresholds, visualize correlations,

and observe the consequences of decision-making in a controlled environment,

The contributions of this work are: an interactive dashboard for risk-oriented project

analytics and a pedagogical model for integrating AI supported dashboards

into PM education.

Deep Neural Networks (DNNs) achieve strong performance

in semantic segmentation for robotic perception but remain vulnerable

to adversarial attacks, threatening safety-critical applications. While ro-

bustness has been studied for image classification, semantic segmenta-

tion in robotic contexts requires specialized architectures and detection

strategies.

We propose a framework for detecting adversarial attacks using pre-

trained ResNet-18 and ResNet-50 models. Our method leverages ad-

vanced feature extraction and statistical metrics to distinguish clean

from adversarial inputs. Experiments demonstrate its effectiveness across

various attacks, offering insights into model robustness. Additionally, we

compare network architectures to identify factors that enhance resilience.

This work supports the development of secure autonomous systems by

providing practical detection tools and guidance for selecting robust seg-

mentation models.

Children in today's world are used to studying through pictures, and the use of visual media in the classroom frequently improves their learning. This article describes a system that helps preschoolers learn the Latin alphabet's letters.

Given the advantages of using the idea of gamification in the classroom, an entertaining and captivating game was developed to assist students in drawing connections between letters and pictures of animals. Children can use the application, which is built on convolutional neural networks, to learn animal names more quickly by connecting dots on the screen to draw letters.

Through the use of different programming languages, the effective training of a letter recognition

model and the development of an intuitive graphical user interface, it has been demonstrated that a practical and attractive application has been created, which has been well received by children.

It has been shown that a useful and appealing application has been developed, which has been well-received by children, through the use of various programming languages, the efficient training of a letter recognition model, and the creation of an intuitive graphical user interface.

In order to make the application more kid-friendly and portable, a Raspberry BI was utilized together with a display in order to show letters generated from the application and an audio system for instructions.

This paper describes the change in teaching practice when delivering Medical Electronics course to the students from two faculties: students enrolled to Applied Electronics specialization from Faculty of Electrical Engineering and students enrolled to Biomedical Engineering from Faculty of Product Design and Environment at Transilvania University Romania. In previous years there was a lack of interaction between the groups of student from these two faculties so the teaching staff has decided to introduce the problem-oriented project work as summative assessment tool alongside the written exam in a safe learning environment which encouraged creativity. The activities included in the group projects followed the KASH model (Knowledge, Attitude, Skills, Habits) because employability represents a combination of knowledge, skills, attitudes, and personal attributes of graduates enabling them to contribute effectively to the workplace and society progress in general.

CRISPR/Cas-based biocomputation has enabled the design of advanced logic circuits analogous to CPUs. However, the scalability of these systems is limited by the lack of formal verification frameworks. Synthetic genetic circuits (SGCs) fail unpredictably due to resource competition (metabolic load), off-target effects, and crosstalk. The design-build-test cycle is slow and incapable of guaranteeing robustness. This article explores the use of Evolutionary (NEP) and Genetic (NGP) Processor Networks as a unified Bio-CAD (Biological Computer-Aided Design) framework. A conceptual mapping is presented where NGP processors model logic gates, strings model gRNAs and mRNA, and formal filters (rc_s, rc_w) model binding specificity. We argue that the inherently parallel and distributed architecture of NGP, and its demonstrated ability to solve NP-complete problems in linear time, directly reflects the massive parallelism of molecular computation in vivo. This approach would allow for the formal design and verification of fidelity, robustness, and scalability of biocomputers before their costly wetware synthesis.

The purpose of the exercise is truly multifaceted. We’d like to teach the students about digital twin technology, stepper motors, H-bridge, limit-switches, hardware hookup, calibration and PLC programming. The pedagogical framework for this student lab activity is grounded in problem-based and student-active learning, with segmentation as a key component. Segmentation involves breaking down complex tasks, concepts, or problems into smaller, manageable units that are logically connected. Each segment targets a specific sub-skill, enabling learners to master one part before progressing to the next.

The authors have used this assignment in their teaching for 3 years, about 150 students/hand-ins. Results are that the students seem to appreciate the modularity of the assignment, even if it is challenging. Furter the students appreciate the joy of seeing the wing physically move as intended from their own coding. As teachers we see that the students seem quite motivated by the assignment. they ask good questions and learn “the hard way” that a software simulator and hardware can be quite different.

In essence, the narrative about the fighter jet wing serves to give students some motivation and context it could be any stepper motor application.

This project addresses the power overload problem faced by a glass bottle manufacturing enterprise that expanded to five automated production lines and a glass-melting furnace. The existing power system, originally designed for smaller operations, could not withstand the transient current surges—five to eight times the normal level—generated when multiple high-power lines started simultaneously. Manual time-based scheduling was used to stagger production start-ups, but it lacked stability and real-time adaptability.

To solve this issue, a multidisciplinary team of instructors and students from Rui De International College collaborated with the company to design and implement a PLC-based intelligent control system for dynamic power load management. The system adopts a three-tier architecture comprising a data acquisition layer, a control layer, and a management layer integrating PLC logic control, sensor data acquisition, and HMI-based visualization. Four core modules were developed: intelligent staggered start, automated control logic, data analysis, and protection and alarm mechanisms.

Practical application demonstrated that the system effectively eliminated overload-induced shutdowns, stabilized current fluctuations, and improved reliability and production efficiency. With a total cost of approximately RMB 80,000–100,000—far below the RMB 500,000 required for transformer replacement—the solution proved both technically and economically advantageous. The project also exemplifies how vocational colleges can collaborate with industry to deliver innovative, low-cost smart manufacturing solutions while enhancing students’ applied automation skills and instructors’ practical teaching competence.

The 21st century industrial landscape is undergoing an unprecedented transformation, driven by artificial intelligence (AI), sustainable digitaliza-tion, and the rapid evolution of smart building technologies. These devel-opments are redefining how knowledge is generated, transferred, and ap-plied within both industrial and educational environments. This paper ex-plores the synergy between smart industry and education in the era of In-dustry 5.0 — an emerging paradigm emphasizing human-centric, resilient, and sustainable systems. Through an interdisciplinary analysis of European and global case studies, including initiatives in Germany and Ukraine, the study identifies effective models that integrate AI, smart infrastructure, and sustainability principles into educational ecosystems. Empirical evidence from university–industry collaborations demonstrates that such integration enhances technical competence, promotes creativity, and builds resilience within future-ready workforces. Drawing on the works of Koeberlein-Kerler (2021, 2023), who pioneered the concept of hybrid learning ecosys-tems and living laboratories for engineering education, the paper substanti-ates that the merging of smart industry and education is essential to achiev-ing the goals of sustainable innovation, digital inclusion, and professional adaptability.

The rapid emergence of generative AI is reshaping educational practice by requiring teachers to renegotiate professional roles, competencies, and class-room routines. This study examines how Estonian teachers understand and adopt AI, drawing on survey data from 210 teachers, reflections from 40 trainers and IT-education experts, and a stakeholder focus group conducted in 2025. The analysis deliberately centers on teacher readiness rather than student experiences, treating it as a prerequisite for how AI is framed, regu-lated, and enacted in classrooms. Using the Kübler–Ross change curve and the Concerns-Based Adoption Model, the study maps teachers’ emotional readiness and adoption stages across school types. Results show rapid growth in digital and AI competencies but uneven, largely individual-driven adop-tion, with persistent uncertainty around ethics, academic integrity, and im-plementation. The findings indicate that national initiatives must align more closely with teachers’ foundational needs, emotional readiness, and organiza-tional conditions to enable responsible AI integration by 2026/27.

The OSKA 2024 monitoring report forecasts for Estonia that by 2027, Estonia will need at least 40,200 ICT specialists, 1.5 times more than in 2020. Cybersecurity and data analysis (including AI) tasks are often carried out by professionals with other job titles (e.g., IT managers in small firms), so official statistics underestimate their actual numbers. Estonian Vocational Education and Training (VET) curricula were developed in 2014, and discussions for a current update began in 2024, with the aim of implementing changes in 2026. Since 2014, the technology has undergone significant changes. For example, cybersecurity threats have evolved rapidly, and generative artificial intelligence (AI) technology appears to already be impacting how students approach modern learning. The Estonian president also announced a new 2025 program called AI-Leap, whose goal is to implement AI into education. As the school's overall goal is to impact students' knowledge, skills, and values, these topics should be integrated into the official curriculum sooner rather than later. Currently, the goal is to implement the Digital Competency Model in VET education; however, it has been unclear which version to use, as 2.0 does not include AI, and 3.0 has not been announced yet. Additionally, several cybersecurity and AI-related competency models are available, including the UNESCO AI Competency Framework for Students, the European Cybersecurity Skills Framework Role Profiles, and the ACM/IEEE-CS/AAAI Computer Science Curricula 2023, which can be helpful in delivering a higher-level approach to curriculum updates.

This paper explores the European Union’s Artificial Intelligence Act (EU AI Act), which entered into force in August 2024. The Act is examined through the lens of economic com-petitiveness, focusing on its implications for AI companies operating within the EU. The EU AI Act presents a risk-based classification system for AI systems—ranging from minimal to unacceptable risk—and imposes strict obligations for companies regarding high-risk and general-purpose AI models. While the minimal risk is unregulated, the limited risk AI systems are subject to lighter obligations.

From a comparative perspective, the study contrasts the EU’s precautionary, ethics-driven approach with the more innovation-centric models of the United States and China. By exploring national industry cooperation and policy makers response, using the case study of Sweden’s AI innovation ecosystem, this paper analyses how European states can support innovation and counter the effects that the EU AI Act has on the agility, scalability, and global competitiveness of European AI startups and SMEs. It highlights growing concerns from European leaders who warn that the Act’s complexity, vague definitions, and rapid implementation timelines could lower EU’s competitiveness.

Based on a quantitative survey of 413 Romanian respondents, this analysis reveals a significant "AI Readiness Paradox". While individuals feel moderately prepared for AI's integration into their work, an overwhelming 77% believe the nation as a whole is not sufficiently informed. The findings highlight widespread anxiety about job displacement, particularly for office and administrative roles, which are seen as highly exposed. Furthermore, AI is currently valued more for the personal "comfort" it provides (37%) than for productivity gains, which is reflected in a low willingness to pay for AI services, as 67% of users engage only with free versions. A sense of fatalism is also prevalent, with 35% of respondents believing they will have no choice but to accept AI's pervasive integration into their lives.

TThe past decade has marked a decisive transformation in educational environ-ments, driven by rapid advancements in digital technologies and accelerated by the COVID-19 pandemic. Since 2016, broadband infrastructure, learning man-agement systems, and immersive instructional tools such as virtual and augment-ed reality have redefined teaching and learning processes worldwide. The sudden shift to remote instruction in 2020 further intensified reliance on digital platforms, fostering hybrid synchronous–asynchronous learning modes in the post-pandemic period. Within this context, engineering education—and particularly Computer-Aided Design (CAD) instruction—has undergone significant peda-gogical and technological evolution.

This study examines the development of CAD education over the last decade by conducting a bibliometric analysis of literature indexed in the Scopus database be-tween 2016 and 2026. Using a systematic keyword search and RIS-file extrac-tion, the dataset was processed in VOSviewer to generate keyword co-occurrence networks and overlay visualizations. The analysis identified 828 relevant publica-tions and revealed twelve thematic clusters reflecting pedagogical foundations, digital design tools, emerging technologies, artificial intelligence integration, and professional training. Dominant research themes include engineering education, design instruction, 3D modeling, and Industry 4.0-aligned digital competencies. Emerging topics, such as open-source platforms, benchmarking, machine learn-ing, and data-driven design, indicate increasing convergence between CAD edu-cation and advanced computational methods.

The findings highlight a research transition from software-centric approaches to-ward holistic, technology-enhanced pedagogies that promote digital literacy, ex-periential learning, and interdisciplinary design. However, gaps remain in linking CAD competencies to broader educational transformations such as lifelong digital skills, competency-based learning, and sustainability-driven design practices.

This paper presents a modeling environment for global human-centric digi-tal strategies that emphasize technologically enhanced production. Building on Industry 4.0 foundations, it explores how AI, collaborative robots, digi-tal twins, and sensor-driven simulations can be leveraged to optimize ergo-nomics, productivity, and economic efficiency in work processes. The ap-proach integrates automated analysis tools, intuitive drag-and-drop inter-faces, and a comprehensive robot library to enable skill-based human-robot task allocation and realistic, safe simulations. Designed for education, the system fosters exploratory, problem-based learning, aligning cognitive psy-chology principles with hands-on experiences to enhance understanding of complex cyber-physical systems. By combining technological, ergonomic, and didactic perspectives, the method supports the acquisition of Future Skills, promotes active learner engagement, and prepares students for the challenges of the human centered industrial paradigms.

CONTEXT

Modern military field operations face critical bandwidth constraints due to reliance on satellite uplinks, tactical radio systems, and the need for low electromagnetic signatures. Despite these limitations, legacy network traffic including advertisements, telemetry, background updates, and tracking scripts consume substantial portions of available bandwidth, negatively impacting mission-critical communications. While existing research has explored Software-Defined Networking (SDN) approaches for agile routing and load balancing in constrained military environments, a significant gap exists in solutions targeting traffic filtering at the DNS level to reduce overhead and enhance security posture. This study addresses this gap by examining DNS-based traffic optimization as a complementary approach to existing military networking strategies, building upon proven civilian applications of DNS filtering technologies that have demonstrated bandwidth savings of 10-30% in non-military contexts.

PURPOSE OR GOAL

This research investigates whether deploying Pi-hole—an open-source DNS-level filtering solution—on low-power Raspberry Pi hardware within military field networks can effectively minimize non-essential traffic, reduce latency, and augment security in bandwidth-constrained operational environments. The study seeks to answer the primary research question: Can DNS-based filtering provide measurable improvements in bandwidth utilization, network performance, and security posture when integrated with enterprise-grade routers in simulated military field conditions? The motivation stems from the critical need to maximize available bandwidth for mission-critical communications while simultaneously reducing security vulnerabilities associated with advertising networks, tracking services, and malicious domains. This work aims to demonstrate a scalable, cost-effective solution that complements existing SDN capabilities and can be rapidly deployed in tactical environments where every transmitted byte impacts operational effectiveness.

APPROACH

The study employed a controlled deployment methodology using Raspberry Pi Model 3B hardware (selected for its low power draw of approximately 5W and rugged potential) integrated with an ASUS RT-AX1800U enterprise-grade router in a simulated field environment. The Pi-hole DNS filtering solution was configured as the primary DNS resolver for all client devices, with security enhancements including DNSSEC validation, query rate limiting, and curated military-specific blocklists. The network topology utilized a hub-and-spoke architecture with Cat-6 wired connections, DHCP configuration assigning Pi-hole as primary DNS, and QoS rules prioritizing mission-critical traffic. Data collection focused on four primary metrics: total bandwidth utilization, external DNS query frequency, web page load times, and simultaneous connection counts. Performance was measured by comparing network behavior with Pi-hole active versus inactive states across multiple simulated mission scenarios. Security effectiveness was evaluated by testing detection and blocking capabilities against known malicious domains from military threat listings. The deployment incorporated network segmentation via VLANs to isolate mission traffic, and comprehensive logging enabled detailed analysis of DNS query patterns and potential security incidents.

Integrating advanced technologies into educational environments is essential for training students in industrial automation. In this context, the Polytechnic School of Engineering of Ferrol (EPEF) has incorporated Phoenix Contact PLCnext technology, thanks to its membership in the EduNet - International Education Network. PLCnext technology represents a significant evolution from traditional controllers, expanding their capabilities beyond the languages defined by the IEC 61131 standard. This versatility allows for the integration of modern languages, facilitating the development of more advanced applications adapted to the current challenges of industrial automation. In the educational context, PLCnext offers a hybrid environment that promotes practical learning, experimentation, and the acquisition of cross-cutting skills in programming, control, and data management. The main objective of this work was to design and implement a comprehensive data acquisition and control system for laboratory level plants, using PLCnext controllers as the system's decentralized cores. The aim was to demonstrate the viability of this technology in educational environments by enabling remote control, SCADA visualization, and data logging. In addition, the flexibility of PLCnext to integrate tools such as MATLAB/Simulink in a laboratory environment was demonstrated.

Abstract. This article presents a new framework designed to improve image interpretation in four high school science disciplines through the use of open-weight language models optimized for resource-constrained environments. It also makes a significant contribution to the field of educational AI. While most previous studies focus on proprietary or hybrid models, evaluated using benchmarks such as VisioMath, ScienceQA, SceMQA, or MMMU, our study proposes an alternative

approach. We optimized four open-weight models—Mistral-7B, LLaVA-1.5-7B, Kosmos-2, and Qwen2-VL—selected for their lightweight architecture, low memory footprint, and robustness. The optimization focused on low-rank adaptation (LoRA) and 4-bit quantization in a resourceconstrained environment, specifically to improve the understanding of scientific images. The results of Bench Low, distinguishing between multiple-choice (MC) and free-response (FR) questions, reveal significant variations depending on the type of task, model, and discipline. LLaVA-1.5-7B stood

out in particular for its superior performance, achieving near-perfect accuracy on multiple-choice questions (96.9%) and the highest score on open-ended questions (69.1%), with an overall average of 79.3%. In contrast, Kosmos-2 achieved the lowest results (40.8% overall), with high variability between disciplines, reflecting uneven coverage of training data. For reproducibility purposes, the source codes are fully available on Github: https://github.com/mouazmikail/Bench-Lowv1/tree/master

A novel system is presented that bridges computational fluid dynamics (CFD) simulations with immersive extended reality (XR) environments by enabling tactile perception of flow phenomena through haptic wearables such as gloves. The newly created conversion pipeline from industry standard fluid simulation software to standalone XR devices is described. In the XR environment, users can dynamically adjust visualization styles (e.g., particle tracing) and simulation playback speeds, allowing flexible exploration of flow behavior. Additionally, the system generates haptic feedback by mapping simulation data to per-finger forces on haptic gloves, where each finger independently perceives localized flow data. A key focus is a low-latency rendering pipeline that synchronizes haptic output with user hand movements, ensuring spatial consistency and tactile fidelity across all fingers. By bridging the gap between simulation software, immersive visualization, and high-resolution tactile feedback, this work establishes a scalable framework for making fluid dynamics tangible in XR.

Remote monitoring and control (RMC) operations, built upon IoT and IIoT systems, enable real-time supervision and actuation across smart environments such as industrial facilities and other remote engineering applications. These systems leverage interconnected devices and advanced data analytics to enhance operational efficiency, safety, and decision-making processes. However, their reliance on light-weight communication protocols with limited built-in security features, such as MQTT, exposes them to cyberattacks that can compromise reliability and safety. This paper investigates the operational impact of such attacks on an IoT/IIoT-based RMC system and proposes a hybrid Edge–Fog deployment of machine learning (ML)-based intrusion detection systems (IDS) to strengthen resilience and ensure continuous operation.

Project-based learning is an effective pedagogical approach in contemporary computer science education, integrating theory with practical skills through real-world projects that cultivate critical thinking, problem-solving, and collaboration. At the XXXXX, students engage in an Informatics Project after foundational coursework in computer science, programming, and mathematics. Traditionally focused on full-stack software development—including front-end, back-end, databases, and containerized deployment—the project was expanded for the 2025/2026 winter semester to incorporate embedded hardware integration via the XXXX remote laboratory platform. This integration addresses industry demands for hybrid software-hardware competencies essential for cyber-physical systems, IoT, and smart industries.

This paper examines the pedagogical benefits and technological challenges of embedding remote hardware access within a PBL-based full-stack software development course. It investigates how hands-on interaction with embedded microcontrollers enhances students’ mastery of asynchronous software architectures, microservice orchestration, containerization, and communication protocols like MQTT and RESTful APIs. By integrating physical hardware controllers into an otherwise software-centric curriculum, this approach bridges academic preparation and industrial multidisciplinary competencies, preparing students for roles in Industry 4.0 and IoT development.

The Informatics Project’s core task challenges pairs of students to collaboratively develop a multiplayer quiz application spanning multiple technology layers: a Bootstrap and TypeScript web front-end, a Java Vert.x backend, and MariaDB for persistent data. The system supports real-time multiplayer interaction through both web-based and hardware game controllers remotely accessed via the XXXXX platform, which employs ESP32 microcontrollers. Controllers communicate with the backend using scalable, industry-standard asynchronous protocols. The semester begins with a kickoff introducing project constraints and a preconfigured Docker Compose environment containing essential microservices and demonstration setups to establish baseline communication patterns. Students manage source control and collaboration in GitLab using structured branching and issue tracking, mirroring professional workflows. Remote lab access enables continuous hardware testing from any location, supplemented by on-campus openLab sessions for mentoring and technical support.

Preliminary results demonstrate that integrating remote hardware into PBL enhances student engagement and motivation by providing tangible interaction with physical devices absent in many software-centric programs. This setup helps students grasp asynchronous communications, containerized service orchestration, and embedded system programming more deeply. The remote lab mitigates traditional physical accessibility barriers, supports inclusive and hybrid learning modalities, and fosters teamwork, version control discipline, and professional workflow adherence. Early feedback indicates increased confidence in developing complex distributed systems and hardware-software co-design, with a greater appreciation for interdisciplinary engineering.

In conclusion, expanding full-stack education with remote lab hardware access through PBL creates a scalable, adaptable, and effective teaching framework bridging software and embedded systems education. The XXXXX remote lab offers hands-on hardware feedback in a controlled yet remote environment, significantly lowering experiential learning barriers while supporting geographical and educational inclusivity. This model offers promising opportunities for broader adoption to enhance project authenticity in engineering curricula, emphasizing structured project management and continuous mentoring to optimize learning. Future work aims to rigorously quantify learning outcomes, scale to larger cohorts, and refine hybrid curricula aligned with evolving workforce needs in smart industries and connected system development.

This paper evaluates an adapted Double-Diamond learning framework designed to foster AI and XR literacy in higher-education design studios using an immersive metaverse environment. The five-week sequence com-bines analog sketching, prompt-based generative AI, and validation through extended realities, emphasizing human authorship, critical reflec-tion, and responsible creativity. A mixed-methods approach was em-ployed: quantitative measures included an adapted Motivated Strategies for Learning Questionnaire (MSLQ) and the Utrecht Work Engagement Scale—Student version (UWES-S9) administered pre- and post-intervention, while qualitative reflections supported interpretation of be-havioral and attitudinal changes. Results indicate consistent improve-ments in AI/XR-related motivational and learning strategies and in aca-demic engagement. On a 0–10 scale, MSLQ scores showed a substantial increase in self-reported strategic use of AI and XR tools, whereas UWES-S9 indicated stronger patterns of vigor, dedication, and absorp-tion. We discuss implications for evidence-based smart education and propose AI/XR-literacy-by-design as a replicable model for creative dis-ciplines.

The Colegio Nacional de Educación a Distancia (CONED) is a Costa Rican institution that is a pioneer in Latin America in offering secondary education through distance learning. The institution's educational process is predominantly conducted in virtual environments, with face-to-face attendance being reserved for the administration of summative evaluations in various branches throughout the country.Due to its considerable enrolment, CONED lacks sufficient physical space and trained staff to carry out face-to-face experimental practices with large groups. In this context, remote laboratories (RLs) emerge as the appropriate alternative, given their capacity to facilitate uninterrupted access to online experiments. This feature of RLs serves to democratise learning opportunities, catering to a diverse student population in terms of age, ethnicity and geographic location. In order to overcome the limitations of the CONED modality, the incorporation of Remote Laboratories is allowing the execution of real remote experiments. To enhance experimental practice and offer more immersive experiences, a pilot was conducted with the Microscopy LR, developed by the Remote Experimentation Laboratory of the Universidad Estatal a Distancia in conjunction with LabsLand, winner of the GOLC 2025 award. This resource employs a real microscope that gives students the experience of working with biological samples. The aim of the article is to analyse the educational experience of CONED's seventh year science students on the use of this laboratory. The study was developed using a mixed methodology to analyse students' perceptions following utilisation of the Remote Microscopy Laboratory. An experimental activity, which was included in the evaluative components of the subject, was implemented, this activity comprised seven sections, which allowed students to be guided through a process of exploration and progressive analysis, fundamental to the development of scientific skills by encouraging observation, interpretation, and the integration of theory and practice. An expert-validated questionnaire was designed and administered to 35 seventh-year science students for data collection. The first section sought to collect information on the academic profile of the students. The second section comprised nine Likert-type statements. In addition, 5 open-ended questions were included to obtain qualitative information aimed at identifying areas for improvement. The results showed that remote laboratories incorporated into CONED's science classes bring the experimental component of science closer to a heterogeneous population such as the one studying in this modality, which can strengthen the development of scientific skills in this population. It is recommended to integrate artificial intelligence to support the use of remote laboratories in distance education and to facilitate student learning when the teacher is not available.

Cybersecurity training struggles to keep pace with the rapid emergence of new vulnerabilities and attack techniques. Traditional methods rely heavily on manual setup, static exercises, and delayed integration of real-world threats, leading to outdated and ineffective learning experiences. Meanwhile, the accessibility of Large Language Models (LLMs) has enabled attackers to automate vulnerability discovery and exploit generation, emphasizing the urgent need for adaptive and continuously updated defensive training tools. This paper proposes HackEd, an automated system that uses LLMs and Retrieval-Augmented Generation to create and personalize Capture-the-Flag challenges. The system operates in two phases: challenge creation, where data from multiple knowledge bases is processed to generate vulnerability-based exercises packaged as Docker containers, and experience tailoring, where an adaptive LLM tutor monitors user performance and provides real-time hints and feedback according to each trainee’s skill level. A human-in-the-loop evaluation was conducted to assess technical validity and pedagogical value. The generated challenges were reviewed for realism and educational effectiveness, while controlled student experiments measured engagement and learning outcomes. Results show that LLMs, when combined with structured knowledge bases, can generate relevant challenges with minimal human intervention. Moreover, the adaptive tutor significantly improved learners’ comprehension, motivation, and retention by offering individualized guidance during problem-solving. Technically, the system reduces the cost and time required to update cybersecurity training content. Pedagogically, it enhances engagement and inclusivity by dynamically adjusting difficulty and feedback. Overall, HackEd demonstrates the potential of integrating LLM-driven automation into cybersecurity education, providing a scalable, efficient, and continuously evolving framework to prepare the next generation of cybersecurity professionals.

This analysis is based on in-depth interviews with the creators of AI products, including Romanian developers and tech entrepreneurs. A central finding is the "Creator's Conundrum": developers use AI to become hyper-productive but are acutely aware that the same technology could devalue or render their own profession obsolete, especially at junior and mid-levels. They view AI as a sophisticated "meta-tool" for amplifying their skills—a "thinking partner" or a "tutor"—but exhibit a deep technical skepticism regarding its accuracy, leading to a consistent professional practice of verification. The study also identifies the "Creator's Burden," a heightened awareness among developers of the systemic risks their creations might pose, including massive environmental impact and the potential for large-scale disinformation.

Drawing from in-depth interviews with Romanian entrepreneurs and managers across diverse industries like healthcare, beauty, and agricultural technology, this study uncovers a "Leadership Paradox". Leaders are optimistic about using AI as a tool to amplify their own productivity, but they express deep pessimism about its long-term societal effects, such as the deskilling of future generations and a collective loss of human ambition. The primary barriers to AI adoption are identified not as technological, but as fundamentally human and cultural, including a deep-seated resistance to change and a social stigma that frames AI use as "cheating". The findings also point to a complex workforce "reconfiguration" that threatens skilled but procedural white-collar roles, not just manual labor.

In the high-stakes world of motorsports, tire performance is often the deciding factor between victory and defeat. As racing technology continues to evolve, IR matrix sensors and modular software are transforming how teams monitor and manage tire conditions. These technologies provide a more precise and data-driven approach to tire monitoring, allowing teams to make real-time decisions to optimize tire performance and strategically manage tire wear and degradation. Moreover, with the ability to simulate tire profiles before a race, teams can proactively adjust their strategies, further enhancing their ability to outperform competitors.

This paper presents ProtectU, a multi-model cyber threat detection system designed to safeguard vulnerable users (such as children and the elderly) against phishing, romance scams, social engineering attacks, and malicious URLs. The architecture combines artificial intelligence modules deployed via gRPC microservices and served in Docker containers. ProtectU integrates seamlessly with a user-friendly web interface and supports multilingual data processing (English and Romanian), ensuring accessibility and effectiveness in diverse linguistic contexts.

In contemporary times, the advancement of cutting-edge technology leads to a need for more people to understand complex systems and how to work with them, education is therefore required to advance along with the rise of technology. One such technology that enhances education in particular practical ways is Virtual Reality (VR) and Extended Reality (XR), offering a new angle for researchers and educators to pursue in the ultimate goal of creating new education methods that can be used to help modern day persons better interact with the technology they use in their day-to-day lives and careers.

The paper is devoted to the progress of the Romanian School of Fluid Power Systems developed by the authors in the frame of Online Engineering. The main stages of creating a modern patrimony of knowledge, hardware and software in the frame of the Doctoral School of the Power Engineering Faculty of the National Science and Technology University POLYTECHNICA Bucharest are presented. The assessment of the value of the main achievements of the former PhD students is briefly presented. The scientific and technical contributions can be identified in the industrial impact of more of 50 graduated from this valuable doctoral school.

The main purpose of the paper is the assessment of the manner of promotion the concepts and tools offered by the on-line engineering in the complex field of the power systems which use fluids under pressure. The wide interest of an efficient high education system in this field comes from the need to collect and use of wide interdisciplinary knowledges and skills, gathered by different scientific and technical field as fluid dynamics, electrical drives, sensors and transducers, analogue and digital controllers and communications.

The integration of digital laboratories into engineering education offers a transformative approach to bridging theoretical knowledge and practical application. This manuscript presents the "DigiMatLabExam" project, developed at TU Dortmund University, which incorporates ultra-concurrent remote laboratories (UCRLs) into the Fundamentals of Materials Engineering course and its corresponding e-examinations. By simulating key materials testing methods (tensile test, hardness testing and charpy impact testing) and enabling detailed microstructure analyses, the digital laboratory fosters competency-oriented learning for over 350 students annually. Initial evaluations revealed enhanced student motivation, confidence, and practical understanding of material properties. Furthermore, a structured didactic implementation ensures that students develop critical competencies during lectures and exercises to prepare for summative assessments effectively. This innovative methodology aligns educational practices with industry demands, paving the way for scalable advancements in STEM education.

Higher Education Institutions are not only digitalizing their degree programs but also aiming for a more open and inclusive Higher Education. Among these efforts to increase inclusivity lies the accessibility of degree programs, courses and the university campuses to enable students with disabilities and/or chronic illnesses (SwDI) to participate in everyday university life.

However, digitalization and accessibility are often not planned for together and accessibility gets relegated to an afterthought or quick fix applied in the end. This creates barriers for SwDI to participate as intended. This is even more apparent in STEM laboratories (conventional and digital ones) which are normally not accessible.

The project "Lab4All - Access@DigitalLabs" dedicates itself to researching and improving the accessibility of digital laboratories ranging from Desktop VR-Simulations over Ultraconcurrent Remote Laboratories to real-time Remote Laboratories. Besides legal regulations and guidelines of the WCAG, SwDIs are integrated into all processes from assessing needs, requirements, experiences and expectations to testing the accessibility of selected laboratories to evaluating the success of revising the laboratories to be more accessible.

The fast evolution of Industry 4.0 and smart factory technologies requires a shift in engineering education toward greater integration of hands-on learning with advanced industrial systems. This paper presents two educational experiments aimed at enhancing student preparedness and engagement in robotics and automation. The first investigates the effect of pre-lab virtual training using digital twin software on student efficiency during robotic programming tasks. Results show that prior virtual preparation reduces completion time and supports more effective problem-solving, although learning transfer varies among individuals. The second experiment evaluates student perceptions of the Beckhoff XPlanar platform as a teaching tool for industrial transport systems. Survey results reveal strong appreciation for the educational value of real hardware, despite challenges in programming difficulty and time constraints. The findings highlight the pedagogical benefits of combining simulation-based preparation with hands-on experience in modern robotics education.

In the context of the digital transformation, the clergy functions as a unique spiritual workforce whose mission is the ethical and psychological guidance of communities. Today, the members of these communities ("parishioners") are deeply embedded in a digital ecosystem that generates new and complex existential challenges: algorithmic addiction, social isolation, ideological disinformation, and anxieties related to the intersection of technology and humanity, such as transhumanism. These issues inevitably enter the pastoral sphere, demanding a new set of competencies from priests.

However, traditional clerical training, specific to seminaries and theological faculties, has been slow to adapt to this new reality. Curricula rarely include dedicated modules on digital ethics, internet sociology, or the pastoral implications of Artificial Intelligence. This discrepancy creates a significant competency gap, leaving many clergy unprepared to effectively address the problems their parishioners face in their daily lives.

The primary purpose of this study is to identify and map the key competencies and strategic frameworks necessary for effective pastoral ministry in the 21st century. The research aims to define a model of the "digital shepherd", a spiritual guide equipped not only with theological knowledge but also with the discernment and skills required to navigate and provide guidance in a novel digital landscape.

This goal is pursued by answering the following research questions, based on the analysis of the perceptions of Orthodox clergy in Romania: 1) What new professional challenges does the digital environment pose to pastoral practice? 2) What specific competencies (technical, ethical, communicational) are necessary for priests to effectively guide their communities? 3) What are the competing institutional strategies the Church can adopt in the online environment (e.g., defensive isolation vs. missionary engagement)?

The study employs a qualitative methodology, focusing exclusively on the in-depth analysis of semi-structured interviews conducted with 8 Orthodox Christian clergy in Romania. The sample is diverse, including parish priests, monastics with technical backgrounds, and academic theologians, to ensure a comprehensive perspective on the professional challenges across different segments of the Church.

Thematic analysis of the transcripts focused on identifying recurrent themes related to the professional practice of the clergy. These include pastoral care strategies in the digital environment, debates on the Church's institutional policy towards technology, the personal difficulties of priests in using new media, and, most importantly, the perceived deficits in their own professional training to face these new realities.

The global discourse on workforce readiness for the AI era focuses almost exclusively on technical upskilling and reskilling. Training programs prioritize teaching employees to operate AI tools, manage data, and adapt to automated workflows, treating the human worker as a component to be updated for a new industrial paradigm. This approach, while necessary, is centered on a purely instrumentalist view of human capital, emphasizing efficiency and productivity above all else. This paper argues that such a focus is critically incomplete. It overlooks the profound impact of advanced digital technologies on the inner lives and moral frameworks of workers, affecting their autonomy, ethical discernment, capacity for deep work, and sense of purpose. Current training prepares workers to operate technology but not to resist its potentially dehumanizing tendencies. This study explores this "inner curriculum" of non-technical skills, drawing on unique insights from a community grappling with the tension between ancient tradition and hyper-modernity.

The primary purpose of this research is to identify and articulate the humanistic and ethical competencies required for workforce resilience in an AI-saturated society. The study is motivated by the need to move beyond an instrumentalist view of AI training and to propose a more holistic model that equips workers not only to use technology but to remain its master. This goal is pursued by analyzing the perceptions of Orthodox Christian clergy and laity in Romania. The key research questions are: 1) What are the primary spiritual and ethical risks AI poses to human agency and well-being? 2) How do these risks manifest as challenges within the workforce, particularly regarding critical thinking and interpersonal relationships? 3) What "spiritual skills" or virtues from Orthodox tradition can be translated into a secular framework for modern ethical workforce training, moving from passive consumption to active, value-driven engagement with technology?

The study employs a qualitative methodology, utilizing in-depth, semi-structured interviews with a sample of 8 Christian Orthodox clergy and 6 lay believers in Romania. This two-tiered sample was chosen to capture both the "pastoral-analytical" perspective of spiritual leaders observing broad societal shifts and the "lived-experience" of laypeople navigating these technologies in their daily lives, from young IT engineers to disenchanted seniors. Interview transcripts were analyzed using thematic analysis to identify recurrent patterns. Key themes that emerged include: the tension between technology as a threat versus a missionary field ; the asymmetry of power between tech architects and users, likened to a game of "Mafia" where the uninformed majority is easily manipulated ; the generational divide in digital adoption and perception; and the imperative to "sanctify" technology through intentional, value-driven use rather than fearful withdrawal.

Generative AI is increasingly being integrated into education, offering transformative possibilities for both teaching and learning and tailoring educational content to the needs of individual students. We have chosen to elaborate on questions lying in the intersection of economics, linguistics and innovative engineering, aiming to benefit from automation achievable using AI. In our previous research we established methodological approaches to testing ability of Large Language Models, in particular Google Gemini and OpenAI ChatGPT, to process figurative speech and tasks in mathematics and physics defined in plain English; the present work is a solution for modelling financial projections of technological startups and competitor search. The said projections are based on an array of variables such as research and development cost, operational expenses, monthly increase of the customer count, cost of customer acquisition and so on; the input data is provided partially in numeric values and partially in the form of linguistic variables, the latter are translated into numbers (or weights) which are configurable, and modelling is done. The output of modelling is processed by a large language model to obtain a textual description in English, or any other supported language. This solution shall be used in the training of students of computer science and automation programs to help them master innovative project management and improve their business development skills. At this stage our solution relies on API-based access to 2 commercially available pretrained language model, namely, Gemini version 2.0 and OpenAI version GPT-4o, at a later stage we are considering development of our own language model. Besides, since our students are not English native speakers, generative capabilities of large language models assist them with the language barrier.

Globally, the demand for IT professionals continues to exceed supply, high-lighting the need for early, accessible, and inclusive career guidance. Tradi-tional counseling services often lack the capacity to personalize pathways in-to rapidly evolving fields such as AI, cybersecurity, and sustainable technol-ogies. To address this gap, Tallinn University of Technology (TalTech) de-veloped AI Career Pathfinder (AI-Pathfinder), a conversational AI system that aligns learners' interests with bachelor-level IT programs. Designed for students, adult learners, and high school teachers, the tool uses LLM-powered dialogue to help users identify suitable study options. Based on data from 479 pilot participants, the study shows how playful, conversational in-terfaces can reduce barriers to IT education while revealing hidden biases, age-related preferences, motivational differences, and technical challenges. An unexpected outcome—human-like interaction—illustrates both the promise and risks of AI-based mentoring. The paper contributes to global STEM outreach by demonstrating the dual potential and pitfalls of integrat-ing AI into career guidance.

Educational remote laboratories allow educators and students to perform experiments with real hardware remotely, addressing geographical and financial barriers in engineering education. However, their potential is often limited by the cost and complexity of specialized physical equipment. The REDTAIL project bridges this gap by integrating high-fidelity simulations with real hardware, creating a more immersive and scalable remote lab experience. This paper presents the REDTAIL unified platform and its accompanying software development kit (SDK), which empowers educators to develop custom simulations, utilize a library of default and community-built assets, and deploy them seamlessly within their courses. Development of default simulations shows the tool chain’s capability to create effective simulations with varying levels of complexity and cross-domain communication between hardware and simulations. Furthermore, we report on successful undergraduate classroom deployments, validating the platform's suitability for higher education. Ongoing development and outreach aim to expand the simulation repository and refine the SDK, advancing the platform's role in promoting a more accessible and scalable future for engineering education.

Abstract. Accurate and timely identification of malignant morphology in histopathology slides remains a major challenge and is still largely performed by expert pathologists scanning whole-slide images for nuclear atypia, crowding, and architectural distortion. Conventional deep learning systems can detect cancer-associated patterns but are typically trained in a static “train-once/predict-once” fashion, behave as black boxes, and offer limited support when confidence is low or domain shift occurs. To address these gaps, we present Look, Mark, Explain, a minimal agentic AI framework that combines classical computer vision with large language model (LLM) reasoning to support cancer slide review while keeping the pathologist in the loop. High-resolution H&E images are processed with a contrast-enhanced, marker-controlled watershed pipeline to segment nuclei and extract 27 interpretable features per cell capturing size, shape, chromatin texture, nucleoli, and spatial context. For user-selected nuclei, a multimodal LLM receives a cell close-up, a contextual tissue view, and the feature vector, and returns a cancer-likeness assessment, confidence, and a pathology-style explanation that explicitly links its judgment to the provided measurements and visible cues, without issuing a formal diagnosis. We demonstrate the system on a breast cancer sample with 971 segmented nuclei, showing that population statistics, size-stratified comparisons, clustering metrics, and LLM narratives align with classical cytologic criteria and highlight enlarged, hyperchromatic, irregular cells in crowded, disorganized regions. The prototype uses standard digitized slides without stain normalization and is embedded in an interactive interface for slide exploration and on-demand AI consultation. While not intended for clinical use, it illustrates how a simple, reasoning-first, pathologist-in-the-loop agent can “look, measure, and explain” in a way that complements existing workflows and provides a testbed for future human–AI studies in digital pathology.

In Electrical and Computer Engineering (ECE) curricula, students typically interact with real hardware to gain hands-on experience in dedicated in-person laboratory sections. Remote laboratories are an alternative by allowing students to interact with the real hardware remotely. However, they are often constrained by the cost and space of physical equipment. This paper presents the development of a novel educational remote laboratory for Programmable Logic Controllers (PLCs) that will integrate real hardware with interactive 3D digital twins. Part of Project REDTAIL, this work details the design and implementation of a unified architecture that synchronizes an Arduino Opta PLC with a simulation of a water bottling plant. Our methodology involved instructor interviews to define requirements, followed by the parallel development of the hardware interface, the backend framework, and the 3D simulation. The primary outcome is a functional prototype that validates the technical feasibility of this hybrid approach. By augmenting a single, physical PLC with diverse virtual scenarios, the system directly addresses the scalability limitations of traditional PLC remote laboratories.

In Electrical and Computer Engineering (ECE) curricula, the hands-on laboratory experience is often hindered by issues like limited physical resources and scheduling conflicts; to address these challenges, REDTAIL implements a multi-module approach, which is centralized in the REDTAIL Simulation Repository (RSR). This paper focuses on the RSR, a scalable, open-source centralized hub designed to provide instructors with a dedicated access and deployment point offering lesson plans, exercises, and various other tools. It features a robust web interface that delivers a clear educational experience, connecting the back-end of simulations, and through integration with \LabsLand, seamlessly incorporates access to remote laboratories. Future work will focus on cultivating an active community of educators within the RSR, enabling them to share and customize educational materials, as well as integrating an AI assistant to provide real-time support and facilitate automated lesson customization and scaffolding. The RSR is thus a valuable step toward solving major issues limiting ECE curricula, empowering instructors, and promoting more efficient learning outcomes.

This paper presents the development and pedagogical integration of a Financial and Accounting Education Application (TECBOOKS) designed to bridge the theory-practice gap for industrial engineering students. Students in this field are required to master complex, interconnected concepts across finance, accounting, and production to evaluate real-world projects effectively. Traditional classroom methods often fail to provide the hands-on, interactive experience necessary to internalize the complexities of project evaluation. This critical skill determines the financial viability of investments.

The TECBOOKS application is a modern, modular, single-page application built on React, specifically engineered to function as a powerful, visual, and accessible analytical tool. Its core innovation lies in its capacity for seamless integration with a separate production line/ERP simulator. This connectivity allows the application to automatically retrieve simulated, personalized production data generated by the student’s strategic decisions in the ERP environment.

Once connected, the TECBOOKS dashboard processes this raw data to generate crucial analytical outputs in real-time. These outputs include dynamic financial statements (Income Statement, Balance Sheet), Key Performance Indicators (KPIs), and sophisticated project evaluation metrics such as Net Present Value (NPV) and Internal Rate of Return (IRR), utilizing various statistical forecasting methods. This integrated approach ensures that the calculation of robust metrics, such as NPV, is not merely a theoretical exercise but is grounded in the realistic operational complexities of a simulated production environment.

The modular design also ensures that individual components—like the statistical forecasting or financial statement generators—can be used independently, making the app scalable and applicable to students in diverse academic fields. Although currently in the early stages of institutional adoption, this solution aims to significantly boost student engagement, financial literacy, and overall comprehension of real-world project complexities through a gamified and data-driven learning ecosystem. Future work will focus on empirical evaluation of student outcomes to quantify its effectiveness.

This workshop translates an AI- and XR-enhanced Double-Diamond learning framework into a hands-on experience for educators, instructional designers, and program leaders. Building on a five-week studio project with first-year students from architecture, design, digital art, communication, and music production, the workshop compresses the core phases of the intervention into an intensive, collaborative session.

Participants will experience a mini journey through Discover–Define–Develop–Deliver using a concrete design challenge: creating a small “sensory habitat” rooted in local cultural narratives or sustainability themes. In the Discover phase, participants analyse a short case, sketch initial spatial ideas, and surface opportunities for AI and XR support. In Define, they work in small groups to craft “prompt stacks” that connect design intent with generative AI tools for image or 3D concept generation while explicitly discussing authorship and ethical considerations. During Develop, they explore how extended realities (e.g., AR/VR visualisations, immersive walkthroughs via example media) can be used to validate and refine proposals. In Deliver, groups distil their process into a short narrative and map potential learning outcomes and assessment strategies for their own courses.

The workshop will foreground human creativity, critical reflection, and responsible use of AI, rather than tool demonstration alone. Participants will leave with: (1) an adaptable Double-Diamond activity template for AI/XR literacy, (2) example prompts and reflection questions used with students, and (3) concrete ideas to integrate similar exploratory projects into their own curricula with realistic time and resource constraints.

This workshop introduces a challenge-based learning (CBL) approach for integrating research, design, and development in engineering education through extended reality (XR). Developed at the University of Twente’s BMS Lab, the model addresses the increasing need for multidisciplinary collaboration by teaching students to speak a shared language across research, design, and technical development.

Participants will actively use a one-page project canvas to translate research ideas into structured user stories, requirements, and project management tasks. The session demonstrates how XR tools can serve as a human-centered medium for prototyping, testing, and validating ideas without requiring coding skills.

By working through a live exercise, participants will experience the dilemmas of aligning user stories, requirements, and project management under resource constraints. This practical, interactive format ensures relevance to STE’s focus on human-centered engineering education and highlights how XR can be embedded as a innovation in curricula.

Keeping the body balance is one of the most important objectives of a person in relation to the environment. For this objective are working together the locomotion system, the nervous system and the sensing organs. Because of its importance there are a lot of published papers on the body balance subject.

Most of the published papers are using statistical instruments on some test results. These mentioned results are obtained applying gait equilibrium tests, using test platforms, on groups of human subjects. We preferred to make a model of balance function, using affordable hardware and software instruments and verify it on theoretical cases to be applied in future practice for real cases.

First, we taken the general feedforward mathematical model of three excitable cells, available in other published papers and then applied it to the interconnecting architecture of neurons inside cerebellum. We simulated the mathematical model obtained for the cerebellum, for some theoretical parameters, in MATLAB, resulting in a set of curves for a set of optimized parameters. We realized a simulation schematic in LTSpice with user defined blocks for neurons and synapses, already published in recent papers, and parametrized them using the optimized parameters obtained after the simulations in MATLAB. We simulated disturbances using the simulation schematic in LTSpice and verified that the results were according with the reality, adjusting the parameters.

Our paper gives us the opportunity to confirm some of the known facts about the nervous system but also to meet into discussion some characteristics with which we were not aware of. Improving the simulation elements and adding some calculus power only the imagination could be the limit of the number of simulated situations.

Field Effect Transistors with Graphene (G-FETs) seem to be a serious candidate, as alternative to the traditional MOSFET transistors, benefiting from small dimensions, sometimes devices created on a single atomic layer, but accompanied by higher currents than in silicon, as the literature reported.

Therefore, the G-FETs simulation represents a strong tool to develop this device area. The most reputable device simulation software is Silvaco (Athena, Atlas). How-ever, in the libraries of this software does not exist graphene, as a semiconductor material. However, graphene is a 2-dimensional sheet of carbon atoms, arranged in hexagonal shape, a configuration similar to the diamond. Therefore, the first G-FET transistor simulated in this work will use diamond with all its intrinsic properties.

This work presents the development of a simulation tool, designed to study the impact of physical activity and lifestyle on the health of young adults. The implemented solution produces synthetic physiological signals, such as ECG, EDA, and skin temperature, based on personalised profiles provided by the user. The system integrates a specific regression model for each bi-osignal, geographical route simulation, and interactive visualisations that en-able contextualised analysis of physiological and emotional patterns. The tool was designed to support teaching, prototyping, and research activities in biomedicine. The results were represented in time graphs and interactive maps, allowing observation of the evolution of the signals along the simulat-ed paths. The validity of the signals produced was confirmed through model performance tests, with metrics such as the coefficient of determination (R2) and the root mean square error (RMSE). The results obtained reinforce the physiological plausibility of the simulated signals and the applicability of the system in real digital health contexts.

Obtaining diagnostic material from small pulmonary lesions remains a persistent barrier to early-stage lung cancer diagnosis. Contemporary diagnostic pathways increasingly rely on bronchoscopy techniques and computed tomography (CT)-guided transthoracic needle sampling; however, limited cellularity, obscuring blood, variable smear thickness and other artifacts can compromise both rapid on-site evaluation (ROSE) and downstream classification. This manuscript describes a rapid, ethanol-based liquid preparation that is intended to standardize slide cellular distribution, reduce obscuring background, and generate two complementary cytology slides from minimal sample volume. Immediately after sample acquisition, one to three drops of aspirate or collected material are introduced into a 10 mL ethanol-based preservation vial. Using an F50 filtration device, half of the preserved suspension is processed onto a first slide; the slide is removed approximately 20 seconds after initiation and immediately stained with Diff-Quik or a rapid Papanicolaou (PAP) variant to support ROSE. The remaining suspension is processed in the same manner onto a second slide, which is placed directly into ethanol for subsequent PAP staining and further evaluation. Representative images demonstrate preservation of diagnostic cytomorphology across benign, reactive, and malignant lung cytology categories, including lymphocyte-rich inflammatory backgrounds, macrophage-predominant specimens, atypical epithelial populations, squamous cell carcinoma, and metastatic carcinoma. The workflow is designed to reduce preparation-to-preparation variability and to provide consistent, interpretable slides in settings where tissue acquisition is challenging and time is constrained. Future prospective studies should quantify adequacy, diagnostic concordance, and suitability for ancillary testing within standardized lung cytopathology reporting frameworks.

Background: Ultrasound-guided fine-needle aspiration (FNA) is the cornerstone for triaging thyroid nodules and diagnosing thyroid malignancy. Rapid on-site evaluation (ROSE) can reduce non-diagnostic sampling and repeat procedures while it is resource-intensive, with increased labor demands at a time when many health systems face constraints in pathology staffing and distribution.

Objective: To describe and preliminarily evaluate a label-free computational microscopy workflow for thyroid FNA ROSE that couples standardized cytology smear preparation with quantitative phase imaging (QPI), enabling immediate digital review without chemical stains.

Methods: Unstained cytology smears were imaged with a proprietary computational microscope that reconstructs quantitative phase maps and renders diagnostically interpretable cytology-like images. Matched slides were stained with Diff-Quik for qualitative comparison.

Results: Label-free QPI images preserved key cytomorphologic information (cellularity, architectural patterns, nuclear-to-cytoplasmic relationships and background characteristics) relevant to adequacy assessment and preliminary categorization.

Conclusions: A stain-free QPI workflow on conventional cytology smear preparations for thyroid FNA may support rapid adequacy assessment and tele-enabled ROSE while preserving material for ancillary studies. Further prospective validation with larger cohorts and quantitative performance metrics are warranted.

The article presents a comparative study of two practical implementations of elec-tronic circuits for transferring photovoltaic energy from photovoltaic panels into electrical energy. The initial energy source is sunlight which is converted by pho-tovoltaic panels into electrical energy. First, the efficiency of solar panels repre-sents the amount of sunlight that a panel can convert into electricity and is de-pendent on the type of photovoltaic panels, namely on silicon or polysilicon sub-strate, on the manufacturing technology and is typical for general purpose solar panels in the range from 20% to 25%. The efficiency of the conversion system is defined by the amount of energy harvested from the solar panels relative to the amount of electrical energy delivered to the consumer (load), which can be im-proved by techniques such as the Maximum Power Point Tracking (MPPT) algo-rithm.

The power management unit is an essential assembly in precision applications where reference voltages with increased accuracy are required. The design of these circuits is a challenge, especially when it comes to the automotive field, where supply voltages can vary significantly (between 3.3 and 40 V).

The power management unit consists of a voltage reference, a voltage stabilizer, a bias circuit, a monitoring block and a pre-stabilizer with buffer.

The article presents the design, optimization and implementation of the bandgap voltage reference circuit that must provide a stable and accurate reference voltage, independent of process variations with temperature and supply voltage.

To reduce process variations, bipolar transistors were used because they have low variations with the manufacturing process. The base-emitter voltage of these transistors is approximately 0.6V at room temperature (27°C) independent of transistor dimensions or doping. In contrast, the threshold voltage of a MOS transistor is influenced by the thickness of the oxide layer, its quality, as well as the concentration of impurities present in both the transistor channel and the pol-ysilicon gate. For the implementation of the circuit at the schematic and layout level, the Virtuoso Schematic Editor and Virtuoso Layout Suite programs provid-ed by Cadence were used.

In a context characterized by declining enrollment in technical degree programs and a shortage of interest in engineering vocations, the Guitar STREAM initiative (School of Engineering of Bilbao, University of the Basque Country, UPV/EHU) emerges as an innovative educational response. This project integrates the STEAM disciplines (Science, Technology, Engineering, Arts, and Mathematics) with a strong emphasis on sustainability, engaging students in engineering through the construction of electric guitars and the design of related infrastructures. In doing so, students experience a hands-on, motivating learning process that fosters collaborative work. The initiative not only aims at the acquisition of technical competences but also at the development of professionals who are aware of global challenges and the social role of engineering. Moreover, it has been consolidated as a vocational guidance model that connects both university and pre-university students with the appeal of engineering applied to real-world and creative contexts

The rapid evolution of drone technology has created new opportunities for education, particularly in fostering interdisciplinary, hands-on learning within STEM and STEAM fields. This paper presents DRONAISSANCE: Skies to Classrooms (DS2C), an Erasmus+ Cooperation Partnership in School Education that aims to empower teachers and students to integrate drones as pedagogical and experiential tools in the classroom. The project responds to European priorities on promoting STEM excellence, advancing digital readiness, and supporting teachers’ professional development. It addresses the lack of pedagogical frameworks and training on drone integration by developing two main outputs: the European Drone Educator’s Toolkit and the European Drone Education Platform. These resources equip educators with the technical, legal, ethical, and methodological knowledge required to implement drone-based learning activities across disciplines. The project also seeks to engage underrepresented groups, particularly girls, in STEM through inclusive practices and collaborative learning environments. Through transnational training, local implementation, and the organization of Drone Awareness Weeks and Drone Festivals, DS2C promotes innovation, equity, and sustainability in European school education. The paper discusses the project’s design, methodology, expected impact, and contribution to the European Drone Strategy 2.0 and the European Green Deal by fostering a digitally skilled and environmentally conscious generation.

The Engineering graduates often enter the workforce unprepared for the cross-disciplinary, agile, and AI-augmented demands of Industry nowadays. This study investigates the misalignment between academic engineering curricula and industry expectations through a qualitative analysis of three expert interviews and content mapping across ten course syllabi. Results reveal key gaps in business reasoning, interdisciplinary teamwork, and hands-on system integration. Based on these findings, we recommend targeted pedagogical interventions, including embedding AI-enabled tools, simulating industry-relevant design processes, and requiring multidisciplinary collaboration in early academic years. These suggestions support the ongoing development of Education 4.0 frameworks that prepare engineers for digital-era work environments.

The incorporation of virtual reality (VR) technologies into the biomedical field marks a turning point in the way medical devices are conceived, developed, and tested. A VR tool designed for the creation and validation of personalized biomedical devices brings together immersive environments, advanced 3D modeling, and additive manufacturing. This innovative approach allows engineers and healthcare professionals to explore design alternatives, refine details, and predict potential challenges before physical prototypes are produced. The result is a more efficient, cost-effective, and patient-centered process that aligns with current demands for precision medicine.

One of the most notable contributions of VR is its capacity to simulate real anatomical conditions. Designers can virtually place a biomedical device within the human body and analyze how it interacts with different tissues, movements, and physiological parameters. This ability to visualize interactions reduces risks, minimizes errors, and improves overall device performance. Traditional prototyping often requires numerous iterations and significant resources, but VR enables testing at a fraction of the cost and time. By integrating VR with 3D printing, validated virtual models can then be physically produced, bridging the gap between digital design and tangible solutions.

In addition to its impact on design workflows, VR provides powerful benefits for education and professional training. Surgeons and healthcare practitioners can use immersive environments to rehearse procedures, test innovative devices, and become familiar with complex interventions before working with actual patients. This enhances confidence, reduces risks, and leads to better outcomes. Medical students can also gain a deeper understanding of device functionality by interacting with prototypes in a virtual environment, fostering experiential learning and skill development.

However, the implementation of VR in biomedical contexts is not without challenges. The regulatory environment governing medical devices is strict, and while essential for ensuring safety, it can slow the adoption of new technologies. Healthcare professionals may also be cautious about modifying established practices, especially when learning curves are steep. Technical requirements such as high-performance hardware, accurate calibration, and reliable software add further complexity. These barriers highlight the importance of offering comprehensive training and technical support to ensure successful integration of VR systems in clinical practice.

Despite these obstacles, the potential applications of VR in healthcare are expanding rapidly. Beyond design and validation, VR can support surgical training, rehabilitation programs, and patient engagement. For instance, patients can visualize their treatment plans or devices in a virtual space, improving understanding and trust. Rehabilitation can also be tailored to individual needs, with VR environments providing motivating exercises that accelerate recovery. The accessibility and affordability of VR tools are improving, paving the way for wider adoption across medical institutions.

Ultimately, the future of VR in biomedical device design depends on interdisciplinary collaboration. Engineers, clinicians, and researchers must work together to align technological capabilities with clinical needs. Such collaboration ensures that devices are not only technically feasible but also practical, safe, and beneficial in real contexts. By fostering cooperation and innovation, VR-based tools for biomedical design and validation hold the potential to transform healthcare, making it more efficient, personalized, and responsive to patient needs.

The rapid growth of the e-sports industry has transformed it from a niche entertainment activity into a global economic and cultural phenomenon, generating new professional opportunities for young people. However, access to these opportunities remains uneven, with socio-economic and educational disparities limiting participation and employability. The Erasmus+ project E4PROSPERITY – Professional Roadmap and Opportunities for Sustainable Participation in E-sports for the Youth addresses these challenges by creating inclusive, structured, and sustainable pathways for young people to engage in e-sports as a viable career field.

The project brings together universities, local authorities, and e-sports associations from Türkiye, Italy, Greece, and Montenegro to co-develop an innovative E-sports Training Curriculum and Screening Test, complemented by digital learning tools and multilingual e-books. The project’s outcomes include enhanced youth employability, new educational standards for e-sports training, and the creation of a sustainable European network linking education, industry, and public institutions. By integrating sport, digital innovation, and social inclusion, E4PROSPERITY contributes to the EU’s objectives of fostering youth empowerment, reducing inequalities, and supporting innovation and quality in youth work across Europe.

The accelerating green and digital transitions are transforming labour markets at a global scale, creating new opportunities while intensifying existing skills shortages. According to the Future of Jobs Report 2025, employers identify skill gaps as the primary barrier to business transformation, particularly in areas such as artificial intelligence, data analytics, cybersecurity, and energy-transition technologies. Although these competencies are increasingly essential for emerging job roles, many education and training systems face challenges updating curricula, accessing suitable technologies, and aligning programmes with rapidly evolving industry needs. As a result, the demand for workforce-ready talent continues to outpace the supply of graduates equipped with practical and future-oriented skills.

This round table focuses on practical models for effective education–industry collaboration that enable scalable, implementable, and sustainable approaches to green and digital skills development. Moving beyond theoretical frameworks, the session brings together educators, industry representatives, policymakers, and training experts (hybrid format) to share real-world experiences of designing and delivering impactful workforce development initiatives. These include co-created curricula, dual and apprenticeship systems, modular and micro-credential pathways, technology-enabled learning environments (remote, virtual, or smart labs, university makerspace…), employer-led academies, and rapid upskilling programmes targeting both youth and mid-career workers.

The discussion will explore how institutions and companies can jointly forecast labour-market needs, embed industry tools and smart technologies in teaching, and design assessment models that reflect real work environments. Attention will also be given to the challenges of implementation, such as funding, capacity-building, institutional coordination, and ensuring inclusiveness for diverse learner populations.

By the end of the session, participants will contribute to a shared set of actionable recommendations for strengthening the education–industry interface and preparing learners for emerging green and digital occupations. The round table aims to support the development of a resilient, future-ready workforce capable of driving innovation and sustainability in smart industry ecosystems.

This study examines how applicant profiles, prior experiences, and motiva-tion relate to admission decisions and early study persistence in informatics programs at Tallinn University of Technology (TalTech), Estonia (2020–2025). Drawing on Tinto’s integration model, Person–Environment Fit theo-ry, and Social Cognitive Career Theory, it analyzes how technical prepared-ness, institutional engagement, and self-efficacy shape enrollment and drop-out. The study combines qualitative content analysis of 168 applicant portfo-lios (2023–2025) with institutional enrollment and dropout data from 2020–2025, coding portfolios across five domains, including IT preparedness, en-gagement with TalTech, and extracurricular activities. Results show that prior engagement with the university, systematic IT experience, and bal-anced profiles combining technical and interpersonal competencies are the strongest predictors of both acceptance and persistence. The introduction of portfolio-based admissions was associated with reduced dropout by improv-ing alignment between applicants’ capabilities, motivations, and program expectations. At the same time, limited technical preparation and weak self-regulation increased the risk of withdrawal. Overall, the findings indicate that content-focused admissions can serve as an anti-dropout mechanism, with implications for equitable, data-informed admissions design in compu-ting education.

In the fast-evolving world of technology, is hard for education's system to keep the pace. The curricula for a 3-year study program is established at the enrolment of the students. After 3 years ,the students present their bachelor's thesis usually as an application with the topic related to what they studied and the expertise of the supervisor.

In order to observe the student’s approach to the topic of the bachelor thesis, we let the student to choose their own topic.

Nearly a quarter of the students from the promotion of 2025 volunteer to do their bachelor thesis in this way. The supervising was done in a laissez-faire manner. In the first semester, the communication with all 60 students was done via email exclusively. In the second semester, the supervisors organized one online meeting, two times a week.

Regarding the topics chosen by the students, we saw that there were topics that were not studied at the university, like applications using AI large language models or quantum computing technologies.

OpenAI launched ChatGPT in November 2022. In October 2022, the students began the university and finished in 2025. The curriculum of the study program did not contain any large language model topics. In this study we discovered that nearly 20% of the student implemented their bachelor thesis application that use a large language model (either hosted or using API).

The average grade for these students under our supervision was on the same level as the rest of the students.

By their choices, we see that students are more likely to tackle topics that go with the technology trends than remaining only by the supervisor’s proposals. Also, the students are willing to learn beyond university’s curricula in order to fulfill the market’s needs.

There are many myths surrounding women in IT, as well as the reasons for the lower representation of women compared to men. Several theories exist as to why women do not pursue studies in IT. For example, a common belief is that women are not interested in IT, that they are too afraid to study it, that people perceive it as a field for men, and that women are not considered intelligent enough to study it. Even though companies are expecting more women in IT, schools are reluctant to pursue female student candidates, as they fear that diversity will lower student quality and that an increase in women will result in lower timely graduation rates. The goal of this paper is to address some of these issues through the studies conducted among IT students at Tallinn University of Technology. Despite the growing emphasis on workplace diversity, higher education has been slower to respond to these biases. We aim to determine whether it is possible to identify gender-based differences in study motivation, entry pathways, and perceptions of study quality, and to translate these findings into institutional recommendations that can enhance women’s participation and success in IT education.

This paper is based on two institutional studies conducted at the Institute of Information Technology, Tallinn University of Technology (TalTech): a bachelor’s student survey carried out in autumn 2024 and a master’s student survey conducted in spring 2025. The surveys were designed to explore diversity and gender-related experiences among IT students, as well as to understand differences in study motivation, learning pathways, and perceptions of inclusion.

The surveys reveal clear gender- and study-level differences in students’ experiences and motivations in IT programs at Tallinn University of Technology. While both male and female students share a strong intrinsic motivation—namely, an interest in technology and problem-solving—their pathways into IT and perceptions of study quality differ notably.

The outcomes of the studies provide the basis for policy recommendations for TalTech as well as other IT institutions in Estonia and possibly elsewhere. The recommendations also take into account the broader Estonian educational context and existing research. The paper also seeks to align its insights with national and European gender equality policies in STEM education.

With software engineering courses taught in higher education gaining considerable ground in the past decade, teaching related ethical aspects has become an essential part of educating future software engineers. This short paper will be based on the experience of teaching a computer ethics course to a multicultural group of graduate computer science students. The paper will attempt to answer the following question: to what extent does culture influence ethical views of students in relation to different software development aspects?

Observations and conclusions will be drawn based on class reflection activities, questionnaires and empirical observations gathered during classes. The lessons learned during teaching this course can contribute to clarifying the relationship between cultural background and approaches to particular ethical aspects in software engineering. They can provide valuable guidance for future computer ethics classes, contributing to the essential education that future software engineers need.

The complexity inherent in the End Degree Projects (EDP) within engineering programs frequently results in substantial challenges for students regarding project preparation and execution. This also contributes to a heavy workload for the faculty involved in tutoring these works. The presented educational innovation project seeks to address these issues based on developing audiovisual resources intended to support students in EDP preparation and reduce the tutoring burden on teaching staff. This approach aims toward more efficient and sustainable learning outcomes.The primary motivation behind the study described here was to conduct a thorough needs assessment from the perspective of the teaching staff. The goal was to precisely identify the aspects of the EDP process that pose the most recurrent difficulties for students. The information gathered serves the critical purpose of determining what specific types of support or additional resources would be most beneficial, enabling the team to adequately identify and prioritize learning needs to focus the content of the audiovisual resources to be developed. The comprehensive needs assessment derived from the faculty survey establishes the necessary empirical foundation for developing targeted and effective educational support. The identified priorities will directly guide the content, format, and pedagogical design of the upcoming audiovisual resources. By aligning resource development with the empirically verified needs of the teaching staff, the project maximizes the potential for these videos to efficiently support student learning in the EDP process and successfully reduce the faculty’s tutoring load, thereby enhancing the overall quality and sustainability of the educational practice.

This paper proposes a methodology for implementing Artificial Intelligence (AI) in a dairy barn system at CAETEC (Experimental Agricultural Field from Tecnologico de Monterrey) to optimize dairy production, enhance milk nutritional value, and promote environmental sustainability. The project focuses on integrating data science, predictive models, and AI to select the best dairy cows. The core challenge lies in synthesizing vast amounts of data—including cow genetics, feeding behavior, rumination, milking results from a robotic system, and animal welfare—to inform decision-making. The methodology outlines four steps for implementing AI, including sensor identification, system interconnection, continuous training, and data analysis.

Various monitoring systems and data sources are utilized to feed the databases for the AI algorithms, including the Sense Hub system for rumination and health, the DeLaval VMS CLASSIC robotic milking system, the DeLaval activity meter system for behavior and heat detection, and the GREENFEED system for measuring methane emissions. The crucial, non-digital knowledge of the expert (veterinarian) is also digitized and integrated.

The selection of breeding sires, a crucial step in genetic improvement, is presented as a complex decision-making problem that is ripe for AI intervention. The analysis highlights the combinatorial complexity in selecting a stallion, even when considering only four specific genetic types (A2A2, High HHP$, Mastitis resistant PRO, and Polled Genetics) with multiple variables and sire options. The total number of possible ways to select sires, considering badges and characteristics, exceeds 172,000, which necessitates the use of AI to identify the optimal matches.

The paper proposes that a 'cow-match' tool can be developed using data science and algorithms to identify the optimal cow-sire match. This tool is designed to support decision-making in an academic context, offering a practical application for students. The significant number of variables and possibilities demonstrates the immediate need for an AI-powered prompt and decision theory to assist students from different backgrounds, like agronomy (with prior knowledge) and industrial engineering (without previous knowledge), in selecting the best sires based on multiple, often conflicting, objectives (e.g., milk production, cost, reproductive issues, and pollution). The goal is to minimize errors and measure the probability of success in the selection process.

This paper presents a learner-centered pedagogical framework for teaching the Information System Factory (ISF) concept using a flipped classroom and project-based learning approach. Originating in the 1980s, the ISF vision anticipated industrialized, component-oriented information system development, but could not be fully realized due to technological limitations at the time. Advances in cloud infrastructure, AI-assisted development, and low-cost computing now make this vision practically accessible within educational contexts.

The proposed framework integrates flipped classroom pedagogy with hands-on system development activities across the full System Development Life Cycle (SDLC), with particular emphasis on the construction and integration phase. Conceptual content is introduced through pre-class materials, while synchronous sessions focus on coached, collaborative system assembly using open-source tools and single-board computers such as Raspberry Pi. AI-assisted coding sup- ports component selection and exploratory learning, while small-group coaching enables individualized feedback and reflection.

The framework employs Mediathread as a collaborative platform for multimedia annotation and reflection, enabling students to connect theoretical concepts with practical implementation. The ISF concept has been applied in multidisciplinary contexts including Business Administration, Cultural Heritage, and the In-ternet of Things, demonstrating flexibility and adaptability across domains.

Rather than presenting a controlled evaluation study, the paper contributes a design-oriented instructional model informed by teaching practice and formative observations. By integrating flipped classroom principles, project-based learning, and an ISF-inspired development workflow, the framework offers a transferable approach for educators seeking to bridge conceptual understanding and authentic information system development in contemporary higher education.

The digitalization of vocational education requires scientifically grounded approaches to assess learning outcomes resulting from the integration of artificial intelligence, simulation tools, and Learning Management Systems (LMS). In response to the rapid evolution of digital technologies and the need for efficient resource utilization, this study proposes a mathematical model for evaluating the effectiveness of digital tools in training future professionals. The model is based on a system of weighted performance indicators: knowledge quality (Kq), learning time (Kt), student engagement (Ka), and resource expenditure (C). These indicators are synthesized into a composite efficacy index (E), enabling comparative analysis between digital and traditional instructional approaches. The research methodology involved analyzing data from LMS platforms such as Moodle and Google Classroom, as well as AI-based interactive environments. Student engagement was measured through analytical surveys, while instructional scenarios and cognitive load were modeled to assess learning dynamics. Generative AI tools were used to analyze student responses, identify semantic patterns, and refine indicator structures. Data normalization, statistical filtering, and expert-based weighting ensured the model’s validity. The model integrates quantitative metrics (e.g., scores, time, cost) with qualitative indicators (e.g., engagement, cognitive interaction), offering a comprehensive framework for evaluating educational outcomes. Validation was conducted through a comparative study involving two cohorts: one using digital technologies and another following traditional methods. Results demonstrated improved knowledge acquisition, a 22% increase in student activity, reduced learning time, and lower resource consumption in the digital cohort. The composite efficacy index (E) showed a 19% increase compared to the traditional model. This model provides a robust foundation for strategic decision-making in the digitalization of vocational education. Its application supports enhanced training quality, efficient resource allocation, and the development of adaptive, data-driven educational systems.

Digital Twin (DT) technology has been used as a transformative approach for data monitoring and industrial process optimization, especially in older machinery contexts where retrofitting sensor networks is more difficult. This systematic review analyzed 803 publications from 2015 to 2025. The main focus was on sensor placement methodologies for DT applications in industrial machinery. The comprehensive database searches were done using Scopus and Google Scholar databases. We identified and categorized sensor placement approaches into six primary methodologies, including optimization-based methods, machine learning approaches, model-based techniques, information theory methods, rule-based/heuristic approaches, and hybrid methods. Our analysis shows a significant trend in the direction of data-driven and AI-enhanced approaches. For old machinery applications, we found that hybrid approaches combining physics-based models with machine learning show superior performance in scenarios with limited historical data. This work provides practitioners with a comprehensive taxonomy of sensor placement strategies. It highlights a trade-offs between implementation complexity, data requirements, and performance for old industrial systems.

In modern network environments, the exponential growth in data traffic and architectural complexity has made the centralized management of Software-Defined Networks (SDNs) increasingly challenging. Traditional SDN controllers, while programmable, still depend on static rules and human intervention for policy definition and threat mitigation, limiting their responsiveness in dynamic and large-scale deployments. This paper introduces a Large Language Model (LLM)-based agentic framework designed to enhance the cognitive and autonomous capabilities of SDN control planes. By integrating advanced reasoning and contextual understanding from LLMs, the proposed system enables intelligent policy generation, adaptive network reconfiguration, and explainable decision-making in real time.The framework leverages Mininet for virtual SDN simulation, Suricata as a telemetry and event-generation layer, and a multi-agent LLM pipeline for interpreting network events, classifying contextual significance, and dynamically generating flow rules. These rules are executed through an OpenDaylight SDN controller, allowing the network to respond automatically to anomalies, performance degradations, or security events without manual intervention. This closed-loop automation demonstrates how LLM-driven reasoning can transform SDN architectures from reactive, rule-based systems into proactive, self-optimizing infrastructures. The proposed approach supports continuous adaptation, scalability across cloud and IoT environments, and transparency through human-readable reports, marking a step toward fully autonomous and explainable network orchestration.

Collaborative robots promise agility and resilience in manufacturing, yet many deployments fail to deliver tangible relief for frontline employees. Addressing this gap, we present a transparent, LLM-supported procedure for “shift handover to the robot,” which systematically delegates repetitive, routine tasks to an unmanned robot shift while preserving human oversight and value creation during staffed hours. The framework operationalises ambidextrous automation: it protects the exploitation of proven processes while intentionally freeing human time and attention for exploration activities such as improvement, onboarding, observation, and learning. The central thesis is that transparent, data-grounded delegation can convert the potential of collaborative robotics into realised employee relief and stable productivity. This paper summarises context, approach, outcomes, and brief preliminary validation results of the procedure, including “exploration surplus” and practical SME adoption guidance.

At the core is an interactive, checklist-guided decision process that combines fourteen practical criteria with context data from production systems. Evidence includes cycle times, takt adherence, setup and changeover characteristics, error history, scheduling buffers, and environmental, health, and safety constraints. A retrieval-augmented LLM interprets this evidence and proposes candidate tasks for delegation, supporting both cooperative day-shift work and an autonomous, unmanned robot shift. Each recommendation carries a transparent decision trail – prompt-response pairs, data inputs, and scored criteria – that forms standardised handover documentation and enables auditability.

The procedure includes a lightweight task library, exception-handling rules (human fallback, safe stops, and alerts), and scheduling integration so that the robot shift is treated as a governed planning resource rather than ad-hoc automation. We anchor the design in ambidexterity theory by positioning exploration and exploitation as mutually reinforcing over time. Delegating monotony to the robot creates an “exploration surplus”: time and cognitive bandwidth that humans reinvest in improvement and innovation, which, in turn, yields more exploitable routines for subsequent robot shifts. This virtuous cycle is illustrated conceptually through an idealised productivity curve and can be tracked empirically using operational metrics.

We outline measurable targets for repeatable adoption in small and medium-sized enterprises: quantitative indicators include OEE deltas, interruption counts, idle time in unmanned shifts, error rates, and rework; qualitative indicators include perceived workload, trust, and collaboration quality. Expected outcomes are fewer operator interruptions, more stable after-hours execution, and higher acceptance due to clear, auditable decisions. Because the criteria and data interfaces are explicit, the approach is pragmatic to implement and portable across heterogeneous equipment bases. Limitations include data availability and quality, the need for robust safeguards in unmanned operation, and disciplined management of prompts and knowledge bases; we propose mitigations such as minimum data thresholds, watchdog mechanisms, and continuous review of decision logs.

Taken together, the procedure reframes robot handover from a narrow technical integration to a socio-technical management practice. It clarifies which tasks are suitable, why they are suitable, and how their delegation will be evaluated, making automation choices legible to engineers, planners, and operators alike. By turning the robot shift into an accountable planning instrument, the framework helps factories convert potential into realised benefits with traceability, safety, and human-centred relief at scale.

Throughout the entire life cycle of stationary gas turbine units (GTUs), such as industrial gas turbine power stations and gas pumping units, numerous types of tests are carried out. The results of these tests must be documented in reporting materials, including current test records with unit operating parameters over the course of testing, final consolidated protocols for each stage, summary data tables for operational documentation (e.g., passports, technical formulars, etc.), and other related records. The content and format of such documentation are defined by the manufacturer’s requirements and applicable standards, essentially representing standardized forms with a clear structure.In the work authors presenting the easy prompt-base solution for the automated test documentation generation. Implementation of the LLM for the test report generation is allowing to reduce the time for the developing working documentation and reduce the human error.

The automatic generation of Programmable Logic Controllers (PLCs) code serves as a tool for enhancing the intelligence and efficiency of manufacturing systems. PLCs can control a wide range of operations and processes, making them indis-pensable in today's industrial settings. One of the most significant innovations brought in this field of Artificial Intelligence (AI) is automated code generation.

The rapid advancement of AI has significantly impacted software develop-ment, particularly in automated code generation. AI-powered tools as machine learning and deep learning models aim to generate, optimize, and refine code, re-ducing development time and enhancing software quality. AI is becoming an in-telligent assistant, capable of accelerating software development phases. We first explore a comprehensive IEC 61131-3 international standard for multiple PLC programming languages as Structured Text (ST) and Ladder Diagram (LD). ST language is a high-level, text-based programming language used in PLC pro-gramming systems. Due to its structure and syntax, high-level programming lan-guages - like Delphi programming language - offer special readability and main-tainability, making it the preferred choice for developing complex control logic.

One main contribution of this paper is to design a structure for ST high-level programming language with its Delphi-like syntax. This is reflected in the case study where an application is presented that was first virtualized. And the pro-gram for the programmable logic controller that controls the operation of the mo-tors of a conveyor was created, in ST language, by using an AI generation tool that uses the natural programming language method.

This work aims to investigate the available platforms for digital archiving, including open-source solutions, and the implementation of such a solution at Transilvania University of Brasov (UniTBv). The initiative is aligned both with national plans for the development of the digital infrastructure for open science - in particular through the university's participation in the National Open Science Cloud Initiative (RO-NOSCI) - and with European guidelines for promoting transparency in research in South East Europe, reflected in the NI4OS-Europe project. At the national level, there is a lack of a coherent policy on the promotion and adoption of open access practices to research results, which contributes to their marginal presence in the public space. Facilitating open access to scientific information could have a significant impact on the academic community and civil society, stimulating inter-institutional cooperation and enhancing the dissemination of knowledge to citizens.

The metal-oxide-semiconductor field-effect-transistor (MOSFET) is the most used transistor nowadays in both digital and analog applications. With the need for low-power circuits and some analog computation functions, the subthreshold (low current) regime of the MOSFET is also used. Therefore, in modelling one must account for two current domains, over threshold voltage and subthreshold, with two different drain current versus gate voltage formulas. Also, the point of transition from one formula to the other must be determined, therefore a longer calculation time is needed. It is useful to determine a unified model and corresponding formula for MOSFET drain current, so the calculation does not depend on the above or subthreshold value of the gate voltage. This is determined in this work together with the extraction of the MOSFET parameters in a practical case.

Digital systems are becoming increasingly prevalent making a qualified cybersecurity workforce and a foundational "cybersecurity culture" essential in all professional fields. Traditional higher education finds it challenging to adapt curricula rapidly to the practical demands of the cybersecurity field, especially for non-specialist students and not only. This program addresses the gap by providing foundational, hands-on cybersecurity education to students from diverse fields of study, building upon existing knowledge about the necessity for cross-disciplinary cybersecurity competence.

The main goal of the current work is to build the “bridge” between theoretical knowledge and practical skills required by the current cybersecurity practices based on a hybrid approach. The initiative aims to train mixed-discipline students in vital cybersecurity activities, developing a broad cybersecurity culture and preparing them for roles focused on delivering security solutions. The project also seeks to prove the effectiveness of an interactive, practical training platform for this mixed student body.

The anticipated outcome is the successful development of content and training of students in practical cybersecurity solution delivery, in alignment with Equity, Diversity and Inclusion (EDI) principles. It is expected for participants to achieve a measurable increase understanding, knowledge and application of concepts like SIEM and Ethical Hacking, stepping from theoretical background to practical deployment and auditing of security solutions. Additionally, the project anticipates the design of a scalable, validated model for cross-disciplinary cybersecurity education that can be replicated in other institutions and fields of study, fostering a campus-wide cybersecurity culture. By the end of the training program, students are invited to offer cybersecurity services to local community organizations (LCOs).

In the last decades Machine Learning models have become increasingly used in different areas of interest, leading to the increase of computational and energy resources. In this context, sustainability has become a global imperative, and Green AI is currently used to encapsulate the concept of development of high-performing and eco-friendly models.

We propose in this paper a function for the selection of the best prediction model from a model library, based on a criterion which combines accuracy, latency, and carbon footprint.

Recommendation systems and sentiment analysis are two of the most common tasks that require high resource consumption. The purpose of this study is to create a system that is based on these two applications with a user-friendly desktop interface. Unlike a traditional system that involves running a single model for each functionality, this one will have to choose the optimal model based on the real-time values of carbon emissions intensities from Romania region obtained from an Electricity Maps API and other criteria based on user input, also taking into consideration the type of device on which the model is running. This approach represents a step in the emerging direction of Green AI paradigm, which also represents the motivation of this study.

The proposed system is composed of two main components: (1) image-based recommendation system and sentiment analysis. The recommendation system uses the Fashion Product Images (Small) dataset from Kaggle platform. The candidate pre-trained CNN (Convolutional Neural Networks) models are: ResNet50, InceptionV3 and MobileNetV2. The score function for the model selection takes into consideration the following aspects: the accuracy and latency of K-Nearest Neighbors algorithm for image recommendation of the entire dataset, the documented latency of each CNN model, the resolution of uploaded image by user, the emission factor score and the device where the model runs. The image resolution criterion offers a “bonus” if the image is close to the model target or a “penalty” otherwise. The emission factor score value adjusts the model score based on the emission factor value from the API. Higher values disadvantage complex models, while simpler models gain preference. This logic applies to the sentiment analysis component, too, where the utilized models are BERT (Bidirectional Encoder Representations from Transformers) and VADER (Valence Aware Dictionary and sEntiment Reasoner), here a particular criterion being the text length.

Few results of this study are as following: In the recommendation system, it was observed that for large dimension images, if the emission factor from the API exceeds the value of 360 gCO2/kWh, the selected model is MobileNetV2 in support of sustainability and for the sentiment analysis component, if the length of text exceeds 50 characters and the emissions value are over 300 gCO2/kWh, VADER is the selected model.

According to the official documentation of the utilized models, the results align with expectations, reflecting each model’s complexity and dimension. These key characteristics along with CO2 emission data, guide the selection of the appropriate model based on the platform used to run the app and user input.

Teleoperated object handling in hazardous environments poses a risk to operators. The development of drones and robotic systems opens horizons for novel civic applications such as technical inspections, maintenance at high altitudes, special deliveries or interventions in hard-to-reach areas. Previous studies have shown that using a live 3D interface along with safe communication methods, and a modular design can make interventions not only safer, but also more efficient and secure. However, easily accessible and practical prototypes are still needed to demonstrate that these solutions can be implemented at low cost.

This work aims to develop and design a remotely controlled robotic arm. It can be installed on aerial work platforms, including drones. It is a scalable and modular solution that is mitigating the operator risks. Its applicability covers the military and the civilian fields, including operations related to maintenance, inspections, light object delivery, etc. So, our goal is to design, develop and control a modular and simple hardware-software system consisting only of Arduino, servomotors, serial communication/API and a 3D interface which provide accurate control, real-time feedback and operational reliability across such diverse missions.

This paper presents a technology-enhanced, project-based educational framework for teaching intelligent control concepts in engineering curricula. The framework integrates theoretical instruction with computational model-ing and real-time experimentation using a low-cost DC motor platform, ena-bling students to follow the complete control workflow from modeling and simulation to hardware implementation and performance evaluation under realistic non-ideal conditions.

The approach is supported by interactive modeling workflows, MATLAB/Simulink tools, and Live Scripts, and is applied to control prob-lems including proportional-integral-derivative, model predictive, and model reference adaptive control. Representative laboratory case studies illustrate how students implement and experimentally validate their designs on DC motor-based systems and related platforms.

Experimental outcomes and student feedback indicate that the proposed framework improves student engagement and supports the development of practical control skills and conceptual understanding. Overall, the framework provides an accessible and affordable model for intelligent control education, bridging theory and practice through hands-on learning.

Autonomous vehicles, whether mobile (Autonomous Mobile Robots, AMRs) or aerial (Unmanned Aerial Vehicles, UAVs) are extensively studied for various objectives such as inspection, search and rescue, transportation or any other tasks related to human support. Considering that UAVs [1], as well as AMRs become essential devices used for example for delivering packages or mails, and potentially in military applications, it is interesting to evaluate adversarial interactions between them. First, it becomes relevant to evaluate if a drone or a mobile vehicle transporting a package makes the delivery to the correct recipient, and not to another one belonging to a competing delivery company. Second, it is essential to investigate the adversarial interactions between autonomous vehicles applications. In this context, it is mandatory to understand adversarial interactions following game strategies like pursuer-evader dynamics or teams-based games inspired by example like soccer.

The rapid evolution of Artificial Intelligence (AI) has transformed the operation of drone swarms from human-controlled formations into autonomous, decentralized systems capable of collective intelligence. Inspired by natural phenomena such as bird flocks or fish schools, drone swarms act as multi-agent systems where each unit follows simple local rules, producing complex global behavior without centralized control. This paradigm shift unlocks scalability, resilience, and efficiency previously unattainable by traditional single-drone systems. The goal of this research is to analyse how AI methodologies enable intelligent coordination, autonomous decision-making, and cooperative mission execution in drone swarms, demonstrating measurable advantages in efficiency, adaptability, and reliability across military, civil, and humanitarian domains and to prepare for development of architectural solution of a control system for swarms of drones.

Auto-graded unit tests are commonplace in programming education, yet their use as pre-learning diagnostics remains underexplored. This study evaluates a short, optional, pre-course Python diagnostic built around unit tests and immediate pass/fail signals. Although the surface task (imple-menting add(a: int, b: str) -> str) appears simple, the suite demands robust handling of messy numeric inputs (locale formats, fractions, scientific nota-tion, bases, complex numbers, currency symbols, verbal numerals). The in-vestigation addressed three questions: (i) how instant pass/fail affects mo-tivation, confidence, and self-esteem; (ii) how students judge usefulness, fairness, and transparency; and (iii) which design changes they prefer.

Engagement logs from students (N=18) and post-course interviews (N=15) were analysed. One third completed the full suite, while many non-completers still reached advanced cases, indicating exploration beyond su-perficial tinkering. Interviews showed that the test felt authentic and moti-vating when it resembled real development work; however, repeated opaque failures reduced confidence. Participants reported that usefulness and fair-ness would be enhanced by an immediate linkage of tests outcomes to con-crete study steps and by the provision of a minimal, autonomy-supportive hint after repeated failure. Changes in intrinsic motivation followed a con-sistent pattern: curiosity at first ‘fail’, a short confidence boost at first ‘pass’, and a dip during repeated failure that could be recovered with a brief informational cue.

Two complementary typologies are proposed to interpret these responses: a three-profile scheme aligned with self-determination theory and a simple two-axis matrix defined by competence-signal sensitivity and transparency requirement. Practical implications include stating optionality and scope up front, maintaining purely binary early feedback, introducing a single mi-cro-hint after repeated failure, mapping outcomes to syllabus items, and ac-knowledging partial structure with lightweight code-aware indicators.

Simulation-based learning plays a crucial role in robotics education by allowing students to practice complex tasks safely and cost-effectively. Digital twins—virtual replicas of physical robots—foster a deeper understanding of kinematics, control, and automation, yet many open-source simulators lack the reliability and precision necessary for consistent educational use. Commercial platforms like Festo CIROS offer advanced multi-body modeling and inverse-kinematics capa-bilities, but no digital twin exists for the Niryo Ned2 collaborative robot. This study addresses this gap by developing a fully functional digital twin of the Niryo Ned2 in CIROS, including peripheral devices such as a conveyor, object feeder, ultrasonic sensor, and sorting bins. A MELFA Basic program implements color-based pick-and-place operations using inverse-kinematics motion and by integrat-ing digital inputs and outputs. The resulting simulation provides a realistic, in-teractive environment that mirrors physical robot behavior, enabling hands-on experience in programming, sensor integration, and automation workflows with-out physical hardware. The work demonstrates the potential of CIROS as a robust educational platform for digital twins and lays the foundation for future en-hancements, including vision-based and AI-driven automation training.

Neuroanatomy education is challenging due to the spatial complexity of brain structures and the difficulty of three-dimensional visualization in traditional teaching methods. Immersive learning technologies, such as 3D Virtual Learning Objects (VLOs), have demonstrated potential for improving knowledge comprehension in advanced educational settings, aligning with the principles of Smart Education and Digital Twins. This research seeks to evaluate the impact of 3D VLOs as digital twins of neuroanatomical structures on improving the spatial understanding of neuroanatomy students, promoting immersive and active learning strategies.An interactive 3D VLO modeling key brain structures was implemented, integrated into a virtual learning platform. The methodology used was ADDIE for the development of the VLOs and the educational process. An experimental design was applied. Spatial understanding and retention were measured pre- and post-intervention, complemented by usability and learning perception surveys using the TAM model. The hypothesis was tested using the Wilcoxon test, and the alternative hypothesis that VLOs improve spatial understanding of neuroanatomy was accepted. The results identified that students who used 3D OVAs significantly improved their spatial understanding and retention of neuroanatomical content compared to traditional methods. They also reported greater motivation, satisfaction, and interactivity with learning. 3D OVAs as digital twins constitute an effective tool for teaching neuroanatomy and, in Education 4.0, enhancing spatial understanding. This strategy aligns with the objectives of Smart Education and the integration of advanced learning technologies, offering a replicable model for education in health sciences and other disciplines that require complex spatial visualization.

Palpation, a cornerstone of medical examination, relies on the practitioner's tactile acuity and anatomical expertise. This paper introduces an innovative augmented reality (AR) approach to enhance foot palpation training and practice. Traditionally, developing proficiency in palpation techniques requires extensive hands-on experience, which can be time-consuming and inconsistent. To address these challenges, we have developed an AR application for foot palpation training that provides real-time guidance by overlaying palpation zones on a scanned image of the patient's foot.

Our study compared the accuracy of locating the medial cuneiform bone using traditional methods versus our AR app. A paired-samples t-test (n=30) revealed a statistically significant improvement in accuracy when using the AR app (M = 4.55 cm, SD = 0.53) compared to traditional methods (M = 9.28 cm, SD = 3.24, p < 0.001). The mean improvement of 4.73 cm (95% CI: 3.52 to 5.94) underscores the potential of AR technology to enhance anatomical education and improve clinical skills.

The AR application integrates with a wearable device, such as the Head Mounted Tablet (HMT-1), enabling interactive learning experiences for medical students and residents5. This technology not only facilitates skill acquisition but also offers potential applications in remote patient care, such as training visiting nurses to triage homebound patients at risk of diabetic foot complications.

Our findings suggest that AR-assisted palpation training could significantly enhance the learning experience, potentially leading to improved diagnostic accuracy and procedural outcomes in clinical settings. This approach represents a promising advancement in medical education, bridging the gap between traditional training methods and the evolving landscape of healthcare technology.

In this workshop we will discuss with invited and interested participants the actual status of the initiative.

This work presents AR FLOW, an Augmented Reality (AR)-supported laboratory environment developed in Unity Engine, that complements conventional teaching of fluid mechanics by making multi-scale flow phenomena interactive. It aims to promote functioning understanding through a modular, level-based learning design that integrates digital lab scripts, AR visualizations, and context-specific learning tasks.

The resulting environment aims to combine script-based learning and interactive laboratories in quest-based learning, which, using Constructive Alignment and the SOLO-taxonomy, can guide students from superficial learning to a relational and extended abstract understanding (deep learning) of the concepts of fluid mechanics.

The rapid diffusion of generative artificial intelligence (AI) and large language models (LLMs) has intensified the need to foster AI literacy in higher education. Although students frequently use AI tools, their ability to apply them critically and purposefully remains limited, underscoring the relevance of instructional formats that integrate structured AI use. This study examines the rede-sign of a bachelor-level wind channel laboratory in chemical engineering. The traditional cook-book-style experiment was converted into a problem-based learning (PBL) format aligned with the SOLO taxonomy and the European Commission/OECD AI literacy framework. In the revised structure, students autonomously engage with theoretical content, select one of two problem scenarios, and use LLMs for planning, conceptual understanding, data interpretation, and reflec-tion. Pre- and post-tests, motivation and technology acceptance questionnaires, and an instrument on generative AI use were employed. Qualitative results indicate increased student autonomy, deeper engagement with aerodynamic concepts, and improved efficiency compared to the in the years 2024 implementation. Students used LLMs across the first three AI literacy domains engag-ing, creating, and managing consistent with the laboratory’s intended learning outcomes. Supervi-sors observed higher motivation and more purposeful inquiry, while the restructured script re-duced time pressure and improved clarity.

The integration of artificial intelligence (AI) into remote laboratories presents sig-nificant pedagogical challenges that require empirically grounded design frame-works. This study is part of research conducted within the framework of a mas-ter's degree project, and the objective is to develop seven design principles for the use of AI-assisted remote laboratories in chemistry education, derived from a multiple triangulation methodology with an inductive approach. Three comple-mentary studies were conducted: a focus group with teachers, a student experi-ence survey, and an analysis of student-AI interactions in a remote acid-base titra-tion II laboratory implemented at the University of Buenos Aires. Seven princi-ples emerged from a systematic triangulation of data from previous studies. The conceptual structure of the principles, based on generalisable teaching design ra-ther than specific disciplinary content, allows for their transferability to remote la-boratories in various fields, including physics, biology, and engineering, contrib-uting to the democratisation of science education in contexts where traditional re-sources are limited.

Mixed Reality (MR) technologies are increasingly used in engineering education to enable, e.g., virtual laboratory training. Yet, most applications support only individual learning, offering limited opportunities for teamwork and communication. This work-in-progress introduces an MR application that allows both local and remote collaboration, enabling two users to interact with the same virtual setup - either co-located or connected online - while sharing synchronized device states and interaction awareness.

The system integrates technical, linguistic, and cooperative dimensions within one immersive environment. In a joint bilingual course, German and U.S. students engage in problem-based laboratory tasks, alternating between English and German. The multi-user architecture embeds collaboration and language use directly into technical training, fostering engagement, co-presence, and teamwork. By shifting MR from isolated practice toward shared laboratory experiences, this approach provides a transferable model for communicative, collaborative learning in engineering education.

Students in the Biochemical and Chemical Engineering programs at TU Dortmund University often struggle to connect knowledge from individual courses into a coherent understanding. These challenges can be derived from limited constructive alignment and high informational load. For example, the first-semester course Introduction to Biochemical

and Chemical Engineering, which provides an overview of the curriculum, illustrates how integrating content from multiple courses can lead to students feeling overwhelmed. To support constructive alignment, the course has been redesigned with clear intended learning outcomes and active learning strategies, including student-generated exam questions and one-minute papers. Complementing this, ChemSPARK, a virtual reality model of a chemical plant representing an industrial process, provides an immersive, process-oriented framework. By contextualizing content in a realistic industrial setting, it is intended to foster cognitive map development, engagement, and long-term knowledge retention.

This study develops the idea that by visualizing an artistic performance through motion and sound graphs, it is possible to identify errors or tenden-cies that negatively affect the sound and the exact moments when they occur. This provides students with a potential way to reduce practice time and maintain a more balanced and healthier schedule. For this study, recordings from the side-positioned camera were utilized, focusing on the arm handling the bow. The video recordings were analyzed using a video analysis software called Kinovea, commonly used by athletes to evaluate their techniques and movements to enhance performance. However, this software also has appli-cations in biomechanics, as it allows for a detailed analysis of body kinemat-ics to correct movement errors and prevent injuries or medical conditions. The goal was to extract the x and y coordinates of the shoulder, elbow, and wrist throughout the entire movement. This was achieved by placing markers on the respective areas. After obtaining the data, it will be processed in MATLAB to generate the graphs. MATLAB is a widely used programming and numerical computation software, commonly applied in engineering as well as other fields such as economics. The graphs were designed to provide as much relevant information as possible to help achieve the study's objective.

Hypertension remains a highly prevalent cardiovascular condition, strongly influenced by autonomic imbalance and chronic stress exposure. Receptive music therapy has gained recognition as a supportive non-pharmacological strategy capable of modulating autonomic and emotional responses, yet ob-jective personalization remains limited. This study investigated the clinical effects of receptive music therapy in adults with hypertension and applied ar-tificial intelligence–based acoustic analysis to identify musical characteristics associated with cardiovascular improvement.

A total of 134 hypertensive adults were followed for six months, engaging in regular home-based listening, while a highly adherent subgroup additionally participated in supervised sessions with physiological monitoring. Blood pressure and metabolic parameters were assessed longitudinally. Musical stimuli were analyzed using a combined AI framework including acoustic feature extraction, unsupervised clustering, and supervised regression model-ing.

Significant reductions in systolic and diastolic blood pressure, alongside helpful lipid profile changes, were observed exclusively among adherent par-ticipants. Therapeutic musical profiles were defined by slow tempo, harmon-ic consonance, rhythmic stability, and reduced timbral complexity, showing strong associations with parasympathetic activation.

These findings support AI-enhanced receptive music therapy as a viable and scalable adjunct in hypertension management, enabling data-driven person-alization of therapeutic listening strategies.

The paper fits into the new concepts for engineering education in higher and vocational education institutions, including emerging technologies in learning. It approaches interdisciplinary and synergistic elements from nature harmoniously combined with modern technologies to highlight some physical and acoustic characteristics of wood. The novelty of paper consists in the use of electronic systems to highlight the chromatic and acoustic resonance properties of the structures of deciduous wood species (oak (Quercus robur Pall.), field elm (Ulmus minor Mill.), downy ash (Fraxinus pallisiae Wilmott), honey locust (Gleditsia triacanthos L.), ash (Ailanthus altissima (Mill.) Swingle), cherry (Prunus avium L.), plum (Prunus domestica L.). The seven rounds obtained from the listed wood species were mechanically processed and protected against humidity and temperature variations by coatings with transparent varnishes and resins. Their exposure and fixation is achieved using a removable metal structure inspired by biomimetics. The chromaticity of the wood was highlighted with high-power LED-COB light sources designed and executed using 3D printing, using a design inspired by nature. The acoustic resonance of each species is highlighted by playing the audio recording of the seven resonance frequencies corresponding to the mentioned species. These frequencies were extracted by processing the acoustic signals with Fast Fourier Transform (FFT) analysis following experimental modal analysis. From a musical point of view, the sounds obtained represent the pentatonic system frequently used in Afro-American music - bluess. The work highlights this aspect by correlating the sound source with the light indicators placed in the proximity of each woody species. The technologies involved in the paper range from CAD design, CNC machining, finishing and stabilization with advanced materials, 3D printing, testing and signal processing, electrical diagram design and incorporation of electronic lighting and acoustic systems.

This study investigates how AI can act as a pedagogical partner in developing the vocal technique, expressive skills, and interpretative autonomy of opera singers. The research seeks to determine whether conversational AI models can assist in structuring vocal study, providing contextual feedback, and supporting artistic reflection. The motivation arises from practical experience in preparing and performing roles such as Caramello (Eine Nacht in Venedig – Johann Strauss), Ruggero (La Rondine – Giacomo Puccini), and Don Carlo (Don Carlo – Giuseppe Verdi). The study adopts a qualitative approach based on self-observation, reflective dialogue with AI models, and iterative documentation of the vocal study process. Conversations with AI were analyzed to identify patterns of feedback, interpretative guidance, and organizational structuring of practice sessions. The framework also anticipates the use of visual and auditory AI tools, including image and GIF processing, 3D modeling, and audio analysis technologies, to illustrate phonatory mechanisms and to integrate these multimodal resources into future didactic materials.

The paper aims to compare the acoustic and environmental characteristics of anthropogenic and forest areas in order to highlight the parameters that contribute to forest therapy and psychological well-being. Thus, the study was carried out in one of the most famous resorts in Brasov County, with nine measurement points for noise levels, pollution levels, acoustic spectrum and anthropogenic/natural elements in the vicinity being established. All recordings were made on the same day, with weather conditions being approximately constant throughout the study. The results showed that the equivalent noise level decreased by 32% in the forest, compared to the central area of ​​the resort, while the CO2 level was approximately constant, considering that the resort is located in a mountainous area, surrounded by mixed deciduous and coniferous forests. Subsequently, the acoustic recordings were processed in Laboratory Virtual Instrumentation Workbench (LabVIEW) and the frequency spectra were extracted. After listening to the forest acoustic sequences, the level of delta waves increased, and the level of theta and low alpha waves decreased.

Laboratories have played an integral part in science and engineering education for a long time. It makes sense that students have opportunities to put theoretical knowledge in a practical context. Another aspect of laboratory learning is to introduce student to aspects of professional practice, for example in relating to how tasks are completed in an engineering workplace.

The learning outcomes of laboratories have been much discussed, and there are several frameworks to support the analysis of their contribution to the educational objectives, such as the widely cited work of Feisal and Rosa (2005).

As the world is seeing immense change in the way that we interact with laboratory learning activities, and the capabilities of generative AI, it is timely to re-evaluate if our laboratory learning activities are still fit for purpose.

Previous work has shown that laboratory learning activities still have a place in a modern engineering curriculum (Kist et al, 2024).

When traditional laboratory learning activities were developed, it is likely that the learning objectives focused on the procedural nature of the activity. It was assumed that students learned things by completing the activities. As we move to virtual and remote activities, and as AI becomes more capable of performing procedural tasks, it is critical that we regularly review our approach.

In this paper, a framework will be introduced to assist educators in a review of existing laboratory activities, and the development of new laboratory learning activities, to ensure that the learning objectives focus on the real intent of the activity. This will include clearly identifying the purpose of the learning activities and also identifying where AI can be used within the activity.

The goal is to provide a tool to support regular review of learning activities to ensure that the integrity of the learning activities is maintained while fostering the requisite AI capabilities in students.

The development of artificial intelligence (AI) technologies has made it possible to build systems that can continuously evolve, in a way comparable to human education. This type of system can be trained to solve tasks in various fields, gradually advancing from a beginner to an advanced level of performance, autonomously adjusting to their complexity and requirements. The concept AI Virtual Student/Worker (AIV-S/W) promises not only the automation and optimization of processes, the elimination of human errors, the substitution of the human factor, respectively the improvement and personalization of education, offering innovative solutions and high-performance standards in various fields.

On the one hand it is impossible to underestimate the influence of generative artificial intelligence(AI) tools on the educational process. Students and learners by nature will always try to use techniques to limit their efforts to reach a goal. So educators are facing the problem of modifying the pedagogical approaches, so that they could include implementation of AI tools in an ethic way, helping student to reach learning outcomes.

On the other hand, project based learning (PBL) is a well-known didactic format in engineering education as it allows to provide students a reality based project environment. PBL is a student-centered model of teaching and learning by doing (student)projects. Projects are to be understood as complex tasks, based on challenging questions or problems. Projects typically run from 2 weeks to one semester. It involves learners collaboratively, in planning, problem solving, decision making and/or research activities. Students acquire autonomy and responsibility, develop overarching skills and apply knowledge. At the end of the projects students report and/or present their results. Combining the positive outcomes of the use of AI with well-defined project cases, will augment the learning experience, without suffering from the loss of original work for the students.

In the work authors research the possibilities of the retrieval augmented generation for the implementation in project based learning approach for the model based system engineering. This allows to increase the field of application of the existing equipment COCO and improve students motivation.

Although Artificial Intelligence (AI) is transforming the industry, many univer-sity programs prioritize generic AI principles above domain-specific applica-tions. This paper investigates an adaptive course format to overcome the gap to real-world use cases. The suggested concept expands on a two-semester pro-gramming course and introduces students to current AI approaches used in re-search and industry in mechanical engineering. The authors aim to address bu-reaucratic limitations in curriculum reform by focusing on modules that pro-mote flexible, problem-based learning. The course structure will include a lec-ture series with a high-level overview and in-depth workshops on specific ap-plications led by active AI researchers. Students will work on semester-long projects with instructors as clients, which specify the technical requirements to be fulfilled. This dual position requires learners to deal with not only AI tech-niques, but also project management, client communication, and interdiscipli-nary interaction.

The broad representation of AI applications is highlighted by providing a brief overview over the starting topics and goals of challenges set in different me-chanical engineering domains such as product development, control systems and thermodynamics.

This paper explores how smart technologies and artificial intelligence (AI) can contribute to building resilience and professional preparedness among social care workers responding to disasters. Drawing on the theoretical framework of resilience and the “help the helpers” vision, it investigates the ethical and pedagogical implications of integrating AI tools into disaster-related social work education. The study builds on doctoral research on Romanian frontline social care workers’ psychosocial responses to crises, emphasizing the shift from reactive coping to proactive learning in technology-mediated environments. The paper aims to discuss how digital training platforms and intelligent systems can support ethical decision-making, reflective practice, and emotional regulation in high-stress contexts. It proposes a framework for socially sustainable and ethically informed workforce development in the era of smart technologies.

Generative Artificial Intelligence (GenAI) is reshaping the educational landscape. A single GenAI prompt can generate a carbon footprint several times higher than that of a standard internet search. Yet, discussions of environmental sustainability remain limited within GenAI-in-education research, which primarily focuses on usability and ethics. This study examines the perspectives of K–12 students and teachers on the environmentally responsible use of GenAI. A questionnaire comprising 12 Likert-scale items was distributed to participants from 63 schools, yielding responses from 729 students and 66 teachers across 80 schools. The results were quantitatively analysed. The presented Structural Equation Modelling model demonstrated strong performance. Students and teachers generally expressed positive views toward environmentally responsible GenAI practices, although their perspectives differed in six aspects, primarily regarding understanding of computational power demands. The differences were statistically significant, as determined by t-tests.

Integrating symbolic AI planning with robotic execution remains a central challenge in cognitive robotics and robotics education. While automated planning provides powerful tools for high-level reasoning, students often struggle to connect symbolic models with concrete robot behavior. This pa-per presents the design, implementation, and evaluation of an educational workshop that bridges AI-based task planning and robotic manipulation through simulation in NVIDIA Isaac Sim. The workshop was developed as part of an advanced Cognitive Robotics course and combines symbolic plan-ning using the Unified Planning Framework (UPF) and PDDL with execu-tion of manipulation plans by a simulated Franka Emika Panda robotic arm. A Robot Manipulation Language (RML) was introduced to abstract gripper orientations and model pick-and-place actions symbolically, enabling stu-dents to reason about spatial affordances while maintaining executability. The workshop follows a two-stage workflow: modeling and planning manip-ulation tasks, followed by execution and validation in a high-fidelity simula-tion environment. Evaluation results based on student performance, reports, and questionnaires indicate that the workshop effectively supported under-standing of AI planning, robot affordances, and the planning–execution loop. The study demonstrates the educational potential of modern robotic simulation tools for making complex cognitive robotics concepts accessible and actionable.

This article presents nine years of initiatives dedicated to develop Bachelor-level students’ skills for the Industry of the Future. The work highlights how emerging industrial expectations—driven by automation, artificial intelli-gence, IoT, robotics, cybersecurity, and data processing—require new peda-gogical strategies centered on Project-Based Learning (PBL). Since joining the EduNet network in 2017, the University of Orléans has strengthened in-ternational collaboration and integrated professional-grade tools into its cur-riculum. Several major achievements illustrate the impact of this approach: finalist positions in the 2018 Xplore competition, a national victory in 2019, the creation of the world’s first “PLCnext for Bachelor Level” certification in 2021, and a third-place award in the Xplore 2023 contest for an AI-based food-waste reduction project. In 2026, the certification will be updated to re-flect evolving industrial needs and new partnerships. Emerging technologies such as Augmented Reality further enrich future pedagogical developments.

Deep Neural Networks (DNNs) are deep learning models inspired by biological neural networks, used for complex

pattern recognition, visual and auditory data processing, and multidimensional signal interpretation. They have

become fundamental in areas such as image recognition, natural language processing, time series prediction, and

industrial process optimization. Denoising represents a central application of DNNs, with the role of highlighting

hidden structures and increasing the accuracy of analysis. Noise can be introduced by external factors (lighting

conditions, mechanical vibrations, electromagnetic variations) or internal (measurement errors, instrumentation

limitations). Its reduction allows: identifying defects at a microscopic scale, improving the reliability of

communications, ensuring data security, and optimizing visual processing. The goal of using DNNs in denoising is to

exploit their generalization and adaptability to achieve high performance with low data collection costs.

The 4th Industrial Revolution encouraged the emergence of new robotic solutions and advancements to assist human operators and increase efficiency and safety within all industrial environments, such as energy sectors, automotive, aerospace, indoor agriculture and so on. Therefore, research in this direction is fueled by the constant need of intelligent systems that can fulfill specific needs throughout industries, such as real-time assessment and report concerning equipment and environment. For example, the indoor-agriculture industry needs robotic systems able to provide data regarding the crop growth environment and plant healthcare in real-time remotely and on-site. In addition to this, manufacturing and energy industries are in need of assessing in real-time equipment, such as manufacture equipment, control and electrical panels, turbines, power lines to ensure the safety and efficiency of the industrial environment. Thus, the proposed solution is a compact and mobile sensorial robotic platform that can accommodate a wide range of devices, such as RGB, IR, multispectral and stereoscopic imaging cameras, and sensors, such as air-quality, environmental, smoke, Geiger and so on. Another key-aspect is the possibility of remote operation via Internet. In order to accommodate the wide range of integrated sensors an open-source and modular software framework was used, ROS (Robot Operating System). This approach allows multiple sensors and devices to be integrated within a unitary software framework in order to obtain valuable real-time data of the operating environment. Also, sensors can be swapped with minimal software modifications, offering a modular character to the hardware and software system. The proposed solution is suitable in industrial environments, such as manufacturing, automotive, aerospace, indoor agriculture, energy and adjacent environments.

The large-scale deployment of electric buses and trolleybuses in urban public transport requires advanced energy management strategies that consider operational constraints, power-grid interactions, and battery longevity. Digital Twin (DT) technology offers a promising framework to support such strategies, but it depends critically on high-quality real-time data from the physi-cal system. In this regard, this paper proposes a mobile data acquisition (mDAQ) system installed on electric vehicles as a key enabling layer for a Holistic Energy Management System (HEMS). This system collects syn-chronized multiparameter data in electrical, thermal, positional, and operational streams through IoT communication towards DT models that opti-mize charging schedules, energy flows, and preventive maintenance. Its ar-chitecture combines multiplexed sensing, floating differential measurement, Geographic Positioning System (GPS) synchronization, MQTT-based com-munication, and driver-oriented user interfaces. The proposed solution is po-sitioned within an integrated DT-based framework for urban e-mobility and can be deployed in a real public transport operator network.

Accurate estimation of forest structural parameters is essential for precision forestry, carbon stock modelling, and sustainable management. Traditional manual inventories are labor-intensive and often limited by sampling intensity. This paper evaluates the application of the Forest Structural Complexity Tool (FSCT), a robust automated pipeline, for the segmentation and extraction of dendrometric metrics from high-density LiDAR point clouds. We processed data from four distinct plots located in the Postăvaru and Barzava regions using UAV and Mobile Laser Scanning (MLS) data. The study validates the tool's capability to process datasets from varying sensor sources while highlighting current limitations regarding low-resolution Aerial Laser Scanning (ALS). Results indicate successful segmentation across varying stem densities, with a specific focus on the comparative analysis of geometric volume modeling methods.

The Ball and Plate Systems aim to fix a ball at the midpoint of the flat plate by rolling or pitching the plate in the direction of the x and y axes when it moves in any direction on that flat plate due to gravity or an external disturbance source. This study aims to redesign the ball and plate system with a new design and transform it into a laboratory set for use in educational activities.

Ordinary ball and plate systems perform this function using different mechanical methods. The most popular of these is the version driven by ordinary hobby-use servo motors attached to two sides of the flat plate. The primary reason for constructing these training sets in miniature is the use of touch-sensitive plates to detect the ball, and these are not manufactured in any size.

The aim of this study is to provide a more visually appealing training set and to more effectively incorporate servo motors into the theoretical content of the training set. For this reason, the study was conducted with a plate scale of 1000x1000mm. The purpose of the study is to balance the ball in the centre of the plate with different control methods.

Thanks to this study, the ball and plate system was created in a more visual and user-friendly manner. Because the resulting system includes a next-generation controller like PLCnext, it also allows users to test different control algorithms.

In the last decade, the use of automation in the workplace has experienced an exponential rise, enabled by technological improvements, which improve accessibility and enhance adaptability. This shift towards automated workspaces has been made possible by advancements regarding organizational solutions that, simultaneously, facilitate integration into corporate spaces.

The main objective of this project is to improve the performance of organizations, by implementing an internal platform that manages document transfers, both in physical and digital format, maintains the stock of supplies and provides alerts for their unavailability. In order for the application to be efficient and sustainable over an extended period of time, factors such as cost, long-term impact and compatibility with existing processes and practices are carefully analyzed.

A series of algorithms was used, that manages essential functions, such as robot navigation (A*), task customization (Decision Tree) and predictive maintenance (ARIMA). Following the identification of existing deficiencies, an integrated system (InternBot) can be provided as a solution. The InternBot structure has two versions. The exclusively digital version, Digital-InternBot (D-IB), with a reduced cost, offers functionalities such as: personalized task management, priority-based scheduling, anticipation of consumable needs with alerts and automatic orders. The second version, Social-InternBot (S-IB), has a robotic component that takes over time-consuming and repetitive physical tasks, in addition to the digital part.

The results from implementing this complex system, with dual functionality, digital and physical, consist of quick access to information, reduced workload for staff and optimised document flow. Although the system is designed to be efficient, errors such as incorrect document location, confusion between files with similar names or data synchronization problems may occur. These situations are anticipated and will be managed through continuous monitoring and immediate intervention by monitoring staff, thus ensuring the correct and continuous functioning of the proposed smart platform.

The KICK 4.0 research project focuses on the approach of effectively using large language model-based (LLM) generative AI technologies in laboratory-based engineering education. The aim of the project is to enable students and instructors to explore the benefits and limitations of generative AI systems in higher education.

Using a customized Design-Based Research (DBR) approach, an LLM-based AI tool is integrated into an ongoing course, evaluated iteratively, and refined. To this end, criteria for determining the effectiveness of an AI tool in terms of the quality of feedback were developed based on a systematic literature review and a survey of students and instructors. Given the small number of participants, the surveys were evaluated using qualitative empirical educational research methods. The results of this requirements analysis show that these technologies are considered to have great potential in supporting the individual learning process and that feedback quality is central in this.

This article describes the implementation of a teaching and learning scenario with the aim of integrating a Retrieval-Augmented Generation (RAG) chatbot into a laboratory-based course in fluid mechanics. In the Computational Fluid Dynamics (CFD) course, students learn how to use OpenFOAM, a software tool for solving complex fluid mechanics problems. Classes are held in a computer lab. Students work at individual workstations in front of a PC and the instructor demonstrates the relevant calculations and answers students' questions. In this setting, students are provided with a customized RAG chatbot as a “digital AI assistant.” The chatbot is trained to give students feedback that is comprehensible and conducive to learning. The results of this work will highlight factors for instructional design with regard to the effective use of LLM-based AI tools in instructional settings.

As part of the CrossLab project, LabCAR (Lab Constructive Alignment Recommender) was developed as an innovative tool that supports both students and teachers in the targeted design and implementation of learning processes. The system has a modular structure and enables the analysis and optimisation of learning outcomes (LOs), the coordination of learning and teaching activities, and the generation of well-founded recommendations for monitoring learning success. Students benefit from personalised suggestions for preparing for the defined learning outcomes and receive references to suitable learning resources, like digital laboratories. Teachers receive support in developing and adapting teaching methods and assessment strategies to promote the achievement of LOs. As part of the evaluation, various large language models (LLMs) were analysed in terms of their performance in the context of LabCAR. The study showed that all LLMs tested are fundamentally capable of fulfilling the defined objectives. However, significant differences in the quality and precision of the recommendations generated were found. The results underscore the importance of carefully selecting and adapting LLMs to meet the specific requirements of the education sector and ensure effective support for teaching and learning processes.

Cross-reality (XR) laboratories are becoming increasingly relevant in STEM education, enabling students to engage in experimentation beyond the re-strictions of traditional physical laboratory spaces. At the same time, large lan-guage models (LLMs) such as ChatGPT are rapidly entering higher education and raise new questions regarding competence development, learning practices, and educational integrity. The KICK 4.0 project addresses these developments by integrating AI-based natural language processing (NLP) tools into XR-based laboratory scenarios in fluid mechanics, with the aim of fostering both technical and reflexive competencies for a transforming world of work. This paper pre-sents a brief narrative literature review of the use of LLM/NLP tools in STEM learning contexts and analyses these findings through the lens of McLuhan’s Laws of Media (LOM; Enhance, Obsolesce, Retrieve, Reverse). This review suggests that LLMs may enhance creativity, reflection, and problem-solving; obsolesce highly scripted, instruction-driven laboratory formats; and retrieve di-alogical and inquiry-based learning practices. At the same time, risks of reversal become visible, including deskilling, bias, dependency, and ethical concerns. By combining the KICK 4.0 project with a LOM-guided literature analysis, this paper provides a structured reflection on how human–AI collaboration may re-shape laboratory education. The findings underline the need for critical AI liter-acy and balanced integration of LLMs in STEM learning environments.

Museums, as key institutions of non-formal learning, are experiencing rapid digital transformation shaped by emerging technologies. Among these, telepresence robots (TPRs) have attracted growing interest for their capacity to overcome physical and geographical barriers to participation. While immersive tools such as Virtual Reality (VR) and Augmented Reality (AR) are already widely employed to enhance engagement and visualization, the pedagogical and accessibility potential of TPRs remains underexplored.

This study investigates the opportunities and limitations of TPRs in museum education and assesses institutional readiness for adopting emerging technologies. Four research questions guided the work: (1) What is the level of readiness among museum educators and information professionals to integrate digital technologies? (2) How can technologies such as Artificial Intelligence (AI), VR, AR, and TPRs enhance learning processes? (3) What are the key challenges and opportunities for implementing TPRs in museum learning contexts? (4) What is the current technological capacity of museums to deploy robotic and immersive solutions for improved accessibility and inclusion?

A mixed-methods design was applied. First, a survey examined the readiness, experiences, and perceptions of museum educators concerning emerging technologies. Second, on-site observations and semi-structured interviews explored contextual factors affecting implementation practices, especially for TPRs. The survey, which included both closed and open-ended questions, covered digital competences, institutional support, infrastructure, perceived benefits and barriers, and ethical issues. Twenty-one professionals responded, and observations were conducted in twenty-two Estonian museums within the MUIS network.

Findings show high interest among educators in AI, VR, AR, and related tools. Most museums already employ digital solutions such as interactive screens, audio guides, and QR codes. While 81% of respondents expressed readiness to integrate TPRs, only 52% had prior awareness of their educational application. VR/AR tools were more familiar but often limited to temporary exhibitions. TPRs, by enabling remote real-time participation, were perceived as particularly inclusive for individuals with mobility or sensory impairments. Nevertheless, barriers included infrastructural constraints, insufficient funding, and limited staff competences. The study highlights the need for a sector-specific competence framework incorporating AI literacy, immersive design, and digital ethics.

By comparing TPRs with other emerging technologies, the study advances understanding of how museums can support educational objectives, foster inclusion, and drive digital innovation in non-formal learning.

Conclusions and recommendations emphasize that enhancing readiness for TPRs and related technologies requires infrastructural, organizational, and educational measures. Accessibility should be improved through ramps, elevators, barrier-free layouts, and optimized exhibition areas to facilitate robot navigation and visibility. Institutional capacity can be strengthened through designated charging zones, systematic staff training, and awareness-raising workshops. Additional practical skills, such as creating QR codes and audio guides, are needed. Pilot programs, teacher materials, and innovation funding should be developed to support sustainable and inclusive implementation of TPR-based and other technology-enhanced learning solutions.

This study contributes to the discourse on digital transformation in museums by clarifying the educational, accessibility, and organizational implications of telepresence robots and related technologies.

This paper presents the integration of Large Language Models (LLMs) and Physics-Informed Neural Networks (PINNs) into a Computational Fluid Dy-namics course based on deep learning and constructive alignment. The course included a lecture and a computer laboratory, conducted via Jupyter-Hub.NRW, enabling students to access remote Python environments and hardware. A survey of 14 students revealed that LLMs, primarily ChatGPT, were used mostly for programming assistance and debugging, while support for mathematical and physical understanding or generating examples was moderate, and planning activities was rare. Exam results indicate that per-formance differences were driven by cognitive levels as defined by the SOLO taxonomy rather than by AI versus traditional content, suggesting lim-ited intrinsic motivation toward AI. Students actively engaged with three AI Literacy Framework domains—Engaging with AI, Creating with AI, and Managing AI—while the fourth, Designing AI, was addressed through hands-on PINN tasks in lectures and laboratories. These activities exceed secondary education expectations of the AI Framework and represent a tertiary-level deepening of AI Literacy competencies. The findings highlight how AI can support numerics education, enhance learning outcomes, and provide a framework for developing advanced competencies in designing and critically evaluating AI systems in higher education.

Recent transformations in music education highlight the need to balance technical proficiency with the development of 21st-century skills such as creativity, collaboration, critical thinking, and problem solving. Although digital tools have proven effective in supporting engagement and interactive learning, their systematic integration into Romanian higher music education remains limited. This study proposes innovative, technology-enhanced activities for music theory, solfège, and dictation courses, using software such as EarMaster, Auralia, and Tenuto. Adopting a descriptive and design-based approach, the paper outlines strategies that combine traditional musical competencies with transversal skills promoted by contemporary educational frameworks. The anticipated outcome is increased student motivation, creativity, and autonomy through the meaningful use of technology. Overall, the study advocates for a more dynamic and digitally literate approach to music education, aligning with current global trends and promoting pedagogical innovation in the university environment.

The incorporation of digital tools into the traditional singing studio is a con-troversial topic, especially in the context of classical singing. Classical vocal pedagogy requires the direct contact between teacher and student, particular-ly in the early phases of vocal development, to ensure that proper technique is acquired by the student and healthy vocal production is established. Fur-thermore, aspects of vocal emission such as colour, brilliance, and dynamics are strongly influenced by the performance venue, since classical singing does not typically use microphones to augment sound. Moreover, there is al-so the issue of preserving a centuries-old legacy, as attested by vocal treatises dating back to the 17th century. The present study was motivated by the in-creased use of online platforms in the process of vocal education. This raised the question: to what extent can the teaching process benefit from the use of video conferencing tools that require the use of microphones, given that the quality of sung tones is altered when sound waves are converted into electri-cal signals? Furthermore, the authors observed that certain digital tools have been specifically designed to help singers warm up or improve their intona-tion, prompting inquiry into how these emerging tools might impact classical vocal pedagogy.

Context - The integration of AI into music therapy enables the creation of personalized playlists, generation of new music, providing real-time biofeedback and even creation of responsive and adaptive clinical, home-based, and digital therapeutic musical environments. Integration process need to consider hardware and software aspects related to advanced sensing, efficient computing, robust security purposes.

Purpose - The authors perform a critical review of existing publications related to therapeutic use of music for the relaxation purposes by implementing biofeedback loops with consumer-grade biosensing wearables coupled with other devices. Also it is emphasised the requirement for BSc, MSc and PhD students in music therapy to learn about the potential offered by Edge AI in the design and development of emotion-responsive music generation and other therapeutic tools which contain personalised real-time feedback systems.

Approach - Integration process need to consider hardware and software aspects related to advanced sensing, efficient computing and robust security purposes.

Advanced Sensing - Smart wearable devices based on real-time contextual analysis aim to provide a more personalised user experience by analysing environmental data alongside personal biometrics. The precise data acquisition and analysis in real-time is based on the integration of novel multi-sensory fusion techniques that leverage Edge AI so the decision making process is fast and responsive.

Efficient Computing – sustainable AI of Things (AIoT) requires substantial electricity consumption leading to significant carbon emissions. Real-time feedback and improved gesture recognition accuracy are provided by integration of wearable sensors with advanced data processing techniques. Music therapy has an important role in the management of stress and anxiety because the enhanced vascularization of various mesocorticolimbic structures and activation of dopaminergic neurotransmitters regulate the autonomic system, emotion and cognitive function.

Robust Security –The scalable and ethically grounded AI-wearable integration could consider federated learning for privacy, deep learning for noise filtering in EEG data, Generative Adversarial Networks (GAN) for enhanced performance in emotion recognition accuracy, temporal smoothness, and perceptual coherence of emotion-aware music generation

Anticipated Outcomes - The paper aims to provide useful technical information to music therapists, researchers and undergraduate and postgraduate students studying music therapy courses on how edge AI can enhance traditional methods (i.e. lyric analysis, music-assisted relaxation, Guided Imagery and Music, music assisted relaxation etc) by offering new creative possibilities, increasing efficiency, and addressing challenges like personalization, cost, ethical implications of AI use in music therapy.

Conclusions - The latest developments in sensing, computing and security aspects related to the use of edge AI in music therapy enhance the capacity of therapists, students, researchers and other users to generate personalised music, perform real time emotion recognition using long term comfort and ergonomic wearable devices, develop intelligent assistive tools for improvisation, music composition and music theory. etc. This paper provides a foundation for understanding current state of the art for wearable devices used in music therapy, efficient computing solutions and robust security methods. Also it emphasises areas for future research and development related to technological aspects of the interdisciplinary field of music therapy.

Nowadays, artificial intelligence (AI) has a heightened interaction with music, being known to even transform music creation. The implementation of AI is discussed to have both, positive and negative outcome: if on one hand, AI helps musicians to overcome creative blocks, to create new musical experiences and to be an aide in gathering and analysing data, on the other hand, it leads to copyright issues and to the idea of a probable substitution of human being musicians.

No matter the source of music the emotional effect is existent but variable as a function of cultural context, listeners’ background, and musical preference.

The aim of this paper is to use artificial intelligence to study the perception and reception of electroacoustic music through analysing listener data (mainly listening habits) in order to identify patterns and preferences. Audio features, neurological measurements (e.g. EEG Mindwave set), psychological tests and sentiment analysis are used to determine the feelings/emotions led by music.

A look at how Artificial Intelligence reshapes work in Europe shows both hope and unease coexist. Although most people see advantages, nearly two out of three think their jobs could vanish because of AI. A clear majority (84%) insist humans must stay in control. Based on data from Special Eurobarometer 554, this paper explores what lies behind such anxieties through the sociological lenses of thinkers like Marx and Weber. Findings point clearly that unequal risk shapes nervousness: those who feel economically challenged or lack tech confidence report higher stress levels. Still, fear of losing control shows up everywhere, even among those fluent in digital tools, echoing Weber’s unease with the hidden rules of bureaucratic systems. In response, our approach leans on a “Trust by Design” model shaped around the EU AI Act, where openness and shared decision-making help shift attitudes from pushback toward collaboration.

The Future of Jobs Report 2025, published by the World Economic Forum, ranks Artificial Intelligence and Big Data as leading transformative technologies shaping the global labor market, with remote and XR-enhanced technologies implicitly included under “AI and information processing.” yet their distinctive experiential potential deserves explicit attention in engineering education. Considering their growing adoption across industries for modeling, testing, and training - and their potential to complement traditional methods by fostering experiential and collaborative learning - these technologies play an important role in preparing future engineers for emerging roles in the future of jobs. This work explores how remote and XR-enhanced technologies can be integrated as complementary tools that naturally engage students’ curiosity, enhance learning, support inclusive and future-oriented learning. Selected examples from applied projects in mechanical engineering demonstrate practical strategies for integration, and show how these technologies, by improving motivation and conceptual understanding, can expand opportunities for their use while promoting ethical principles such as collaboration, equity, autonomy, accessibility, and lifelong learning, and offering scalability and safety. All these features are strengthened when integrated with AI and Big Data.

This qualitative study investigates the attitudes and behaviors of students in Bucharest, Romania, toward the increasing use of Artificial Intelligence in academia. Conducted through semi-structured interviews and guided by the theory of planned behavior, the research explores how students from diverse fields like sociology, computer science, and medicine perceive and utilize AI as a learning tool. The study addresses the technology's influence on evaluation processes, academic integrity issues such as plagiarism, and the broader ethical implications of its integration. The thematic analysis of the interview data reveals a complex and dualistic perception of AI's role in the modern educational landscape.

The findings highlight a significant divide in how AI is viewed. On one hand, students widely acknowledge its positive impact, primarily valuing it as a powerful tool for efficiency and enhanced learning. AI tools like ChatGPT are consistently used to save time by summarizing long texts, generating project outlines, and quickly locating specific information. This allows students to better manage demanding schedules, which often balance academic responsibilities with employment. Furthermore, AI functions as a personalized tutor, offering alternative explanations for complex concepts and assisting with technical tasks, which is particularly beneficial for students who need supplementary support outside of lectures. It also serves as a source of inspiration, helping to overcome creative blocks by providing a foundation of ideas upon which students can build.

On the other hand, the study uncovers deep-seated concerns about the negative consequences of AI reliance. A primary risk identified is the potential erosion of critical thinking and the promotion of intellectual laziness. The ease of generating complete assignments encourages a "copy-paste" mentality, which circumvents the fundamental learning process and is detrimental to long-term knowledge retention. This dependency raises significant issues of academic integrity, making plagiarism more accessible and widespread. Moreover, students are aware of the technology's limitations, including its tendency to "hallucinate" or produce inaccurate information and fabricated sources, which poses a threat to the quality of academic work.

The research concludes that the impact of AI is not inherent to the technology itself but is contingent on the user's intent and method of application. While responsible use can augment learning, its abuse leads to superficial understanding. A notable gap exists between prohibitive or nonexistent formal academic policies and the widespread, informal use of AI by students. This disconnect points to a need for universities to move beyond simple bans and instead develop clear guidelines for responsible use. Students advocate for pragmatic solutions, including formal training on AI ethics and the adaptation of evaluation methods to test genuine comprehension over rote memorization. Ultimately, the study suggests that AI's integration into professional and academic life is inevitable. The academic environment must therefore evolve from a position of resistance to one of proactive integration, preparing students for a future where human-AI collaboration is the norm and the ability to leverage these tools effectively is a critical skill.

Telecommunications degree programs should transition from "demonstrative" laboratories to development environments in which students design, implement and operate real systems. Integrating edge platforms (Raspberry Pi/ NVIDIA Jetson), Software-Defined Radio (SDR) and digital signal processing tools (GNU Radio/ MATLAB) enables students to transition from theoretical concepts to functional prototypes early in their studies. However, employers increasingly report that graduates have practical shortcomings, such as difficulty integrating theoretical concepts into real-world applications, gaps in hardware and software troubleshooting, inconsistent use of modern tools, unsafe RF operation and inability to reproduce experiments. The proposed approach addresses these issues by constantly emphasizing the connection between theory and practice through incremental spiral projects and clear maturity criteria.

Cybersecurity trainings are based on static, pre-defined "capture the flag" (CTF) tasks and attack blueprints. Although good learning exercises, these practices cannot mimic the spontaneity and worldliness of real cyber threats to Industry 4.0 and cyber-physical systems. The possible disruption brought by artificial intelligence (AI), especially large language models (LLMs) and reinforcement learning (RL), will enable the creation of dynamic and context-based adversarial plans.

The goal of this paper is to determine if an academic framework for training on AI-crafted attacks, instead of training on fixed exercise sets is feasible or not. Our specific interest lies in large language model (LLM)-enhanced reinforcement learning for designing unpredictable adversarial activity (e.g., phishing emails, SQL injection, variations on malware) that adapts continuously against student defences.

We propose a multi-level learning platform, based on LLMs, which constructs attacks uniquely designed for the student's skill level and defence tactics. The framework consists of:

1. Content generation using LLM for natural-language attacks such as phishing.

2. RL-driven adversarial modelling for evolving technical exploits such as injection or malware evasion.

3. Student defence platforms comprise intrusion detection laboratories, secure coding assignments, and network monitoring simulations.

The data gathering will obtain pre/post-competency assessments, defence success rates, and student perceptions. Comparison analysis will measure the disparity between the learning outcome of the AI-based group and a comparison group based on static CTF challenges.

It is anticipated that learners subjected to adaptive AI-generated assaults will exhibit enhanced critical thinking capabilities, quicker detection and response times, and greater retention of defensive techniques. Furthermore, it is expected that they will develop heightened confidence when confronting unpredictable adversaries, indicative of skills that are more applicable to real-world contexts within Industry 4.0. Additionally, the framework offers educators access to scalable, perpetually updated training resources, eliminating the necessity for manually created challenges.