Stölzle, M. Safe Control of Soft Robots: Bridging Physics and Learned Models: Rising Star Abstract. Abstract from 8th IEEE-RAS International Conference on Soft Robotics (2025), Lausanne, Switzerland.
Abstract: The contributions presented in this abstract equip soft robots with the motor intelligence they need to function effectively in human-centric environments. We accomplish this by combining advanced machine learning techniques with physical priors and stability guarantees. By incorporating physical structure into learned models, we enable the use of well-established model-based control methods, ensuring effective, stable, and computationally efficient control. The contributions discussed in this abstract, both current and future, aim to enhance the productivity and effectiveness of soft robotic manipulators (e.g., achieving precise movements at high speeds) while prioritizing safety, compliant behavior, and the development of transparent and inspectable computational intelligence.
More Information
