Philip Tobuschat

Max Planck Institute for Intelligent Systems

Papers

Total Citations

H-Index

About

Philip Tobuschat is a robotics researcher whose work lies at the intersection of machine learning, control, and dynamic manipulation, with a particular focus on robot table tennis. His major contributions center on enabling robots to perform complex, real-time tasks that require both precision and adaptability. In his highly cited 2023 work, "Data-Efficient Online Learning of Ball Placement in Robot Table Tennis," Tobuschat introduced an online optimization algorithm that allows a robot to learn how to return ping-pong balls to a predefined target by optimizing over interception policies—a significant step toward agile, interactive robotics. Complementing this, his paper "Black-Box vs. Gray-Box: A Case Study on Learning Table Tennis Ball Trajectory Prediction with Spin and Impacts" presents a novel gray-box approach that fuses physical models with data-driven learning to predict ball trajectories, accounting for spin and impact dynamics. With over 2 citations each, these papers have already sparked interest in the robotics community. Tobuschat's work is notable for bridging the gap between theoretical modeling and practical, data-efficient learning, making him a rising figure in the field of robotic manipulation and autonomous systems.