Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

Session Overview
SES 1.5: Robotics and Computer Integrated Manufacturing
Tuesday, 27/Jun/2017:
11:20am - 1:00pm

Session Chair: Michele Gadaleta
Location: Aula Q (first floor)

Show help for 'Increase or decrease the abstract text size'

95. How to Deploy a Wire with a Robotic Platform: Learning from Human Visual Demonstrations

Francesca Stival, Stefano Michieletto, Enrico Pagello

University of Padova, Italy

The advent of Industry 4.0 set a new standard in terms of workflow and customization. Agility is a key characteristic for industrial systems and the time needed for introducing a new article should be minimized. Production needs the flexibility to adapt to recurring changes and robotics is a major resource in obtaining this goal. In this paper, we address the problem of deploying a wire along a specific path selected by an unskilled user. An operator teach an arbitrary path by moving in a natural manner a tool deploying the wire through several pegs composing different possible routes. The system recorded the covered trajectories by using a camera network composed by both 2D and 3D cameras. The robot has to learn the selected path and pass a wire through the peg table by using the same tool. The work is part of a more complex project aiming at the development of a learning-based approach for robotized coils winding, to be used in the electric machines manufacturing industry. The configuration selected for our experiments is less demanding with respect to the real industrial environment in terms of movement precision. In fact, the focus of this work is related to correctness and wire deployment. The main contribution regards the hybrid use of Cartesian positions provided by a learning procedure and joint positions obtained by inverse kinematics and motion planning. Copper wire needs to be deployed along the path, some constraints are introduced to properly deal with this non-rigid material without breaks or knots. A learning framework based on Gaussian Mixture Model (GMM) and Gaussian Mixture Regression (GMR) is trained starting from an initial set of examples. Once a first model is ready, an incremental procedure let us introduce new examples starting only from mean, covariance and priors of the components forming the previous model. The projects has been selected for the European Robotics Challenges (EuRoC) and a benchmarking procedure has been created to test the system on a series of metrics validated by a super partes evaluation panel constituted by experts in robotics. Benchmarking tests regard mainly three aspects. The correct deploy of the wire is guarantee is no breaks or knots are present. The number of demonstrations necessary for following the correct peg sequence measures the generalization capability of the system. The time needed for updating the model gives a feedback in terms of learning adaptability. On top of these metrics, we take into account the time needed for computing the robot trajectory starting from the trajectory performed by the human demonstrator. All the parameters tested over performed the targets set during the analysis conducted by our industrial partner.

226. Autonomous Manufacturing of Composite Parts by a Multi-Robot System

Alfons Schuster, Michael Kupke, Lars Larsen

German Aerospace Center, Germany

Aerospace structures require a combination of low weight and high mechanical performance and thus often involve composite materials, e. g. carbon fiber reinforced plastics (CFRP) or fiber metal laminates (FML). The laminate structure involves complex layups, which makes manual production error prone. Today, there still is a lack of innovative production techniques to achieve competitive production rates. Automating this processes demands a smart and flexible, however robust system directly linked to the CAX-chain. We investigated a combination of cooperation industrial robots and computer vision without teaching of the robots.

Commercial products like Delmia or Process Simulate are useful for digital factory planning. Add-ons like Cenit’s Fast Suite extend the functionality towards digital production, but lack the ability of handling huge numbers of cut-pieces. It was shown in previous work that a robot’s target points for gripping and dropping cut-pieces can be derived automatically and subsequently the layup can be carried out autonomously. Also was shown that computer vision strongly improves process accuracy and robustness. The focus of this paper lies on the practical implementation of a smart manufacturing execution system in a multi-robot environment. Major components of the cyber-physical system are the Manufacturing Execution System (MES), the robots and their controllers, one ore multiple computer vison systems for detection of the goods being handled, and a simulation environment called CoCo for collision avoidance.

This work considers pick-and-place processes, which consist of the steps picking, transfer, dropping and post-drop treatment. Our scenario was an airplane skin demonstrator made of dry CFRP sheets. The jig for placing the cut-pieces is half-shell shaped with a diameter of approx. 4 m and a length of approx. 2 m, while the 108 cut-pieces are approx. 1.2 m by 1.8 m and approx. 1.2 m by 0.8 m in size and are provided in a drawer based storage system. For determining where to grip we use one computer-vision system mounted to a KUKA Quantec KR210 R3100 robot. A second, identical robot with identical grippers is operated using the image coordinates of the same camera. Both robots are mounted to one linear axis of 8 m length. The information on which cut-piece to grip, where to grip it from, what is it’s contour and where to exactly move the grippers for gripping and dropping is contained in a proprietary Job Definition File format (jdf), which can be considered a preliminary step to later extensions of the CPS by a production planning and execution agent. A parser converts the generic jdf-information to an action list for each robot comprising setting the tool center point, move the robot, switch the grippers and do the post-drop tacking. The robots receive their actions by the KUKA technology package Ethernet KRL. Time critical movements, like cooperative cut-piece transfer with two robots, are first parametrized and then executed synchronously by the KUKA technology package RoboTeam. Thus, the fully automated, autonomous production of a generic airplane part with multiple robots in a complex environment could be demonstrated.

44. Offline CAD-based robot programming and welding parametrization of a flexible and adaptive robotic cell using enriched CAD/CAM system for shipbuilding

Lucía Alonso Ferreira, Yago Luis Lapido Figueira, Isidro Roberto Fernández Iglesias, Manuel Álvarez Souto

AIMEN Technology Center, Spain

Shipbuilding is usually a handwork process, many shipyard’s facilities are poorly optimized and they haven’t flexibility enough for more complex manufacturing. In 2015, shipbuilding sector emerged with the purpose of dynamizing its R&D environment, developing of advanced manufacturing technologies that make easier the technological evolution of the sector.

The purpose solution is a hyper-flexible welding robotic cell, composed by a gantry of three axes with a six-axes robot assembled. The nine axis are fully coordinated by the robot controller. The dimensions of the three-axes are 5x4x2.5 m, the anthropomorphic robot is a KUKA KR16-2 model. The system is provided with a localization system based on machine vision.

The main topic of this development is the software that help to program this robotic cell in a CAD environment allowing implement welding sequences to an inexperienced programmer: ‘Offline automatic system programming’.

The worker, through the CAD program, is capable of configure the welding parameters and program the robot in an automatic way, generating the robot trajectories module and is responsible for sending it automatically to robotic system, without using the console.

The machine vision locates the part to be welded. This cell has two cameras embarked on the gantry that scan the work space; the software combines multiple images with overlapping fields of view, producing a high-resolution image, locates the part into the created one and calculates its pose – position and orientation –, which is sent to the robot.

The app is embedded in the open source software FreeCAD, in order to make it accessible for Small and Medium-Sized Enterprises. In this program, a workbench was created using Python language. The user selects in the CAD model the joint that wants to weld and, through the interface, the user adds welding parameter’s –e.g. voltage, current, gap…– and location parameters –eg. the workobject or the distance between points-. The welding joint 3D points are extracted from the CAD and the pose of the part to be welded is result from machine vision system. All the data are stored in a configuration file in a XML format. Therefore, this file associates CAD data with process parameters, getting an enriched CAD/CAM file.

With all the data parameterized, the necessary calculations are made to determinate the point coordinates of each joint, and generate a DAT file -KUKA file format that stored the path points-, that is sent to robot. The calculation of the trajectory takes into account the coordinated movement of the nine axes.

All the process instructions are sent to the robot controller from the workbench created on FreeCAD. This is carried through a communication between the robot and the application developed, based on sending and receiving a XML structure.

With this system, manually programmed paths are eliminated by automatic generation off-line with CAD systems. This system is adapted to small batch manufacturing with different parts designs; the system is ‘easy work’ so that inexperienced personnel can use it. In addition, this adaptive cell is more productive than a conventional robotic system.

296. Semantic modelling of hybrid controllers for robotic cells

Mathias Haage1, Jacek Malec1, Anders Nilsson2, Maj Stenmark1, Elin Anna Topp1

1Lund University, Sweden; 2Department of Automatic Control, Lund University, Box 118, 221 00 Lund, Sweden

Programmable Logic Controllers (PLCs) play an important role in integration of hardware and software in industrial robot cells featuring an increasing amount of heterogenous equipment from several vendors. PLCs are also particularly useful for implementing hybrid controllers for the cells. PLCs are defined in the IEC 61131 standard including several programming languages as defined in the 61131-3 part of the standard. In this paper we propose a semantic grounding of the Sequential Function Charts (SFC) notation for specification of PLC programs. Semantic modelling of PLCs allows the use of automatic reasoning methods to accelerate cell setup and (re)configuration, including generation of SFC descriptions. Our semantic grounding is expressed in the OWL semantic language and forms part of our semantic robot framework, called KIF (short for Knowledge Integration Framework). KIF is a set of ontologies and associated tools to ensure interoperability between heterogenous equipment making up a robot cell. We also present a tool set for manipulating SFC instances stored in RDF triple stores reachable through the RDF4J framework. SFC instances may be stored declaratively, analysed, modified (including various forms of composition) and exported into the run-time system for execution. For this last purpose we use the JGrafchart tool. The semantic grounding and tool set are evaluated in a teaching-by-demonstration experiment in a small parts assembly setup featuring a collaborative industral robot, ABB YuMi, where the tool set is used to create and execute SFC descriptions on-the-fly based on data from human demonstrations.

36. A New Model of Modular Automation Programming in Changeable Manufacturing Systems

Tarek Al-Geddawy

University of Minnesota Duluth, United States of America

Manufacturing systems in Industry 4.0 are changeable, smart, connected and more autonomous. The structure of a changeable manufacturing system allows for physical reconfiguration, however, reprogramming controllers has been always performed manually for each new system configuration. The presented model combines different ladder logic codes corresponding to different system configurations, modularizes them and produces smaller pieces of code, which automatically get merged and downloaded to the different system controllers. The model uses Cladistics and Design Structure Matrix (DSM) to prepare the modular codes. A case study of a changeable robotic assembly system is presented.

Contact and Legal Notice · Contact Address:
Conference: FAIM 2017
Conference Software - ConfTool Pro 2.6.119+CC
© 2001 - 2018 by Dr. H. Weinreich, Hamburg, Germany