Research

Prof. Dr.-Ing. habil. Stefan Palis

How can technical systems be designed so that they act reliably, safely, and transparently even under varying conditions? This is the central question addressed by the Chair of Intelligent Automation Systems.

Our research combines model-based and data-driven process descriptions, state-dependent decision-making, and formal safeguards. The aim is to develop automation systems that are not only high-performing, but also robust, explainable, and suitable for reliable use in quality- and safety-critical applications.

The focus is on methodological questions in modern automation: How can nonlinear, only partially observable, and time-varying systems be described appropriately? How can models be adapted or switched during operation? And how can learning-based methods be integrated in such a way that safety and stability remain guaranteed?

We do not examine these questions in isolation from their applications, but in concrete technical domains. Two main areas form the thematic core of the chair: autonomous crane- and robot-based processes, and autonomous production and particle processes.

Our objectives are to:

• investigate automation systems that can be operated robustly and safely even under uncertainty
• systematically combine physical process knowledge with data-driven methods and learning-based approaches
• develop methods for detecting models, states, and regime changes during operation and using them for observation, control, and optimization
• design intelligent systems in such a way that their performance and safety properties are methodologically grounded and practically usable