Tobias Pfaff
Papers
2
Total Citations
16
H-Index
2
About
Tobias Pfaff is a leading researcher at the intersection of machine learning and physical simulation, with a primary focus on learning-based dynamics models for robotic manipulation and rigid-body physics. His work addresses the fundamental challenge of enabling robots to predict and interact with complex physical environments. Pfaff’s major contribution is the development of graph neural network (GNN)-based approaches that learn to simulate rigid collisions and complex physical dynamics directly from data, overcoming the limitations of traditional physics-based models that require full-state information. His 2025 review on learning-based dynamics models for robotic manipulation (9 citations) provides a comprehensive framework for understanding how data-driven approaches can enhance planning and control in robotics. Additionally, his 2022 work on face interaction graph networks (7 citations) introduces novel architectures for simulating rigid-body dynamics among arbitrary shapes, addressing the notoriously difficult problem of complex geometry and non-linear interactions. Pfaff’s research has significant implications for advancing robotic manipulation, enabling more adaptive and generalizable physical reasoning in autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1A review of learning-based dynamics models for robotic manipulation9 citations · 2025
- 2Learning rigid dynamics with face interaction graph networks7 citations · 2022