Jordy Senden

PhD in Robotics | Knowledge Valorisation Officer and Project Manager for the AIMatters project at Eindhoven University of Technology | Owner at Senden Systems Solutions

Engineering Autonomy: How Systems Engineering Can Shape the
Next Generation of Intelligent Robotic Systems

Bio.

Jordy Senden works at Eindhoven University of Technology (TU/e), where he leads the TU/e team in the European AI-Matters project and co-founded the Robotics Technology Center (RTC) together with Wouter Kuijpers. He holds a BSc and MSc in Mechanical Engineering and an EngD (formerly PDEng) in Mechatronic Systems Design, where he first encountered systems engineering, an approach that has shaped his work ever since. He then completed a PhD in robotics at TU/e on world models for robust robotic systems, studying how robots can build and use an understanding of their environment to act reliably in real-world conditions. Within AI-Matters, which establishes Test and Experimentation Facilities to lower the barrier for SMEs to adopt AI and robotics and to make the manufacturing sector more robust, his team focuses on autonomous robots that perform relevant industrial skills. With the RTC, he aims to bridge the valorisation gap between robotics research and industry, accelerating innovation and the uptake of AI and robotics.
He is convinced systems engineering holds the key to making autonomous robots dependable enough for industry.

Artificial Intelligence has significantly expanded the capabilities of autonomous robotic systems. Advances in foundation models, generative AI and learning-based perception enable robots to operate in increasingly unstructured environments and perform tasks that were previously beyond the reach of automation. However, these developments also challenge established engineering practices. As autonomous systems become more capable, ensuring explainability, adaptability, maintainability and verification becomes increasingly difficult.

This workshop explores a systems engineering perspective on the development of intelligent autonomous systems. It argues that scalable autonomy requires architectural abstractions that complement learning-based AI with explicit system knowledge and well-defined functional decomposition. Rather than treating intelligence as a monolithic capability, autonomous behaviour can be realised through the interaction of multiple architectural components, including semantic world models, explicit knowledge representation, task models, reusable skills, resource models, planning, execution monitoring and reasoning.

Special attention is given to the role of semantic world models as the explicit representation of the operational context of an autonomous system. Such models provide the foundation for reasoning, decision-making, monitoring and adaptation, while enabling traceability between system objectives, operational knowledge and system behaviour. The workshop further discusses the separation of Task, Skill and Resource as an architectural principle that promotes modularity, reuse and interoperability across robotic platforms.

These concepts are positioned within a broader Systems-of-Systems perspective, where AI components are integrated as specialised capabilities within a modular system architecture rather than acting as the architecture itself. This architectural approach supports runtime adaptation while maintaining explicit interfaces, verifiable behaviour and lifecycle manageability.

The workshop provides an interactive forum to discuss how classical systems engineering principles—including abstraction, decomposition, interface definition and model-based engineering—can be extended to support the design and engineering of trustworthy autonomous