We create differentiating AI solutions along with concrete lead applications, and ensure to transfer them into Bosch businesses.
AI-Based Dynamics Modeling
Machine learning has been successfully applied on stationary physical systems. However, accurate modeling of dynamic systems is still a challenge. Our goal is to develop statistical, nonparametric and real-time capable dynamics models.
By making inference in probabilistic graphical model-based hierarchical environment models, we provide multi-modal behavior predictions of all agents in any given scene and determine how likely certain situations are to occur.
Control Optimization Through Reinforcement Learning
Automatic calibration and adaptation of controllers is central to the vision of a self-calibrating factory. We intend to exploit modeling knowledge and new exploration strategies for effective learning control.