Dojo - A Differentiable Simulator for Robotics

Dojo - A Differentiable Simulator for Robotics

We present a differentiable rigid-body-dynamics simulator for robotics that prioritizes physical accuracy and differentiability, Dojo. The simulator utilizes an expressive maximal-coordinates representation, achieves stable simulation at low sample rates, and conserves energy and momentum by employing a variational integrator. A nonlinear complementarity problem, with nonlinear friction cones, models hard contact and is reliably solved using a custom primal-dual interior-point method. The implicit-function theorem enables efficient differentiation of an intermediate relaxed problem and computes smooth gradients from the contact model. We demonstrate the usefulness of the simulator and its gradients through a number of examples including: simulation, trajectory optimization, reinforcement learning, and system identification.

Related Papers

People

Taylor Howell
Google Deepmind
Simon Le Cleac'h
Boston Dynamics AI Institute
Jan Bruedigam
PhD at TU Munich
Zac Manchester
Assistant Professor
Last updated: 2022-03-07