Aitor Arrieta // #SISeminarSeries​​ // December 11, 2025

Testing Three Generations of Cyber-Physical Systems ABSTRACT—The complexity of Cyber-Physical Systems (CPSs) has evolved over the last few years. A first generation of CPSs, which are still the prevalent ones in our daily lives, relies purely on hand-coded software, where the functionality is mostly implemented through software (typically in C). As the advances of Artificial Intelligence permitted reliable functionalities, such as perception, prediction and decision making, a new generation of CPSs emerged. These systems integrate learned components, such as neural networks, into the control loop, blending traditional engineered logic with data-driven behavior. This shift greatly expanded their capabilities, yet introducing new testing challenges. Right now, we are witnessing a paradigm shift as Visual Language Action (VLA) models push CPSs into a third generation, where multimodal understanding and high-level reasoning can drive complex behaviors end to end. Instead of stitching together separate perception, planning, and control modules, these systems increasingly rely on large, unified models that interpret visual scenes, follow natural-language instructions, and generate actionable control policies. This promises unprecedented flexibility and autonomy, but it also makes testing to be significantly complicated, making aspects like the test oracle problem a significant endeavor. In this talk I will present our recent works in testing these three generation of CPSs, and I will outline future challenges that remain unsolved. BIO—Aitor Arrieta is a permanent full-time lecturer and researcher at Mondragon University. His research interests include software engineering and software testing areas, mostly related to the application of these to complex software systems such as Cyber-Physical Systems, AI-based systems, and Generative AI models. His work is very industry-oriented, collaborating with different companies to address real-world challenges. Aitor earned his PhD in Software Engineering at Mondragon University in 2017. He works on search-based software testing, the test oracle problem, regression testing, and more. He is involved in several exciting projects including InnoGuard and TRUST4AI. ℹ️ https://si.usi.ch/seminars