Action and interaction in physical autonomous systems

Physical autonomous systems, also known as robots, are a result of a longterm integration of mathematical modeling, software and hardware advances in several fields of technology as well as social sciences. Robots are equipped with various sensors and actuators that enable autonomous interaction with the environment. Similarly to biological systems, the environment provides context for interactions, tools for executing tasks and means for grounding semantics. Central to achieving this is representation and parameterization of multimodal sensory data that enables safe, robust and scalable action generation. This talk presents the mathematical tools used in our research toward this goal.