DiffSDFSim: Differentiable Rigid-Body Dynamics With Implicit Shapes

Abstract

Differentiable physics is a powerful tool in computer vision and robotics for scene understanding and reasoning about interactions. Existing approaches have frequently been limited to objects with simple shape or shapes that are known in advance. In this paper, we propose a novel approach to differentiable physics with frictional contacts which represents object shapes implicitly using signed distance fields (SDFs). Our simulation supports contact point calculation even when the involved shapes are nonconvex. Moreover, we propose ways for differentiating the dynamics for the object shape to facilitate shape optimization using gradient-based methods. In our experiments, we demonstrate that our approach allows for model-based inference of physical parameters such as friction coefficients, mass, forces or shape parameters from trajectory and depth image observations in several challenging synthetic scenarios and a real image sequence.

arXiv comment22 pages, 23 Figures (including supplementary material). Presented 3DV 2021. Project website: https://diffsdfsim.is.tue.mpg.de/

References in corpus (6)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• DiffTaichi: Differentiable Programming for Physical Simulation
• gradSim: Differentiable simulation for system identification and visuomotor control
• Scalable Differentiable Physics for Learning and Control
• Sparse Surface Constraints for Combining Physics-based Elasticity Simulation and Correspondence-Free Object Reconstruction
• Accurately Solving Physical Systems with Graph Learning