Time: Thursday, 28/Aug/2025: 10:30am - 12:00pm Session Chair: Alois Herkommer, University Stuttgart, Germany |
Location: Collegezaal D |
INVITED
Inverse methods for design in nonimaging optics
1Eindhoven University of Technology, Netherlands, The; 2Signify Research
We consider the inverse problem from nonimaging optics: given a source and its light distribution, find the optical surfaces that transform
the light into a desired target distribution. We present models for optical systems with a parallel or point source. The surfaces (lens or reflector) are freeform. Our models are based on Hamilton's characteristic functions and energy conservation.
Designing Fresnel phase elements with differentiable raytracing
KU Leuven, Belgium
Phase elements can enhance the performance of imaging systems while reducing their size, by adding continuous phase gradients on top of the geometrical surface. Compared to their typical implementation as diffractive elements, Fresnel optics can achieve similar performance, while maintaining broadband functionality. This work introduces a methodology to optimize such Fresnel surfaces, resulting in high-performance, compact imaging designs.
An inverse method to compute freeform optical surfaces for generalized zero-étendue sources
1Eindhoven University of Technology, The Netherlands; 2Signify Research, The Netherlands
We present an inverse method to compute freeform reflector and lens surfaces for generalized zero-étendue sources. The initial position and direction of a light ray is parameterized by two source planes and the final position and direction of a light ray is parameterized by two target planes. We use energy conservation to determine optical mappings, and we use the optical path length to derive equations for the optical surfaces. In two numerical examples, we illustrate the algorithm's capabilities to tackle complex light distributions.
Comparison of Phase Space and Quasi-Monte Carlo Ray Tracing for the 3D Compound Parabolic Concentrator
1Eindhoven University of Technology, Netherlands, The; 2Signify Research, Netherlands, The
Phase space ray tracing is an alternative to (Quasi-)Monte Carlo ray tracing in 2D. We introduce a 3D phase space algorithm and apply it to the compound parabolic concentrator. Our results show that phase space ray tracing outperforms Quasi-Monte Carlo ray tracing in 3D.
Hybrid Neural and Deconvolution Approach for Finite-Source Reflector Design
1Eindhoven University of Technology; 2Signify
We present a hybrid method for reflector design with finite light sources, combining a neural-network-based solver with a deconvolution-inspired iterative correction scheme. Our approach addresses the limitations of classical techniques, which often assume idealized point or parallel sources, by solving a simplified problem using a neural network and refining the solution via feedback from ray-traced simulations of the full finite-source system. We demonstrate the effectiveness of our method on a representative example, showing improved convergence toward a prescribed far-field intensity distribution compared to the approximate problem's solution.