Efficient Differentiable Simulation of Articulated Bodies
This work addresses the need for gradient-based optimization in neural networks operating on articulated bodies, accelerating reinforcement learning and control applications, though it is incremental as it builds on existing simulation techniques.
The paper tackles the problem of integrating articulated body dynamics into deep learning frameworks by presenting an efficient differentiable simulation method, resulting in an order of magnitude faster computation and two orders of magnitude reduced memory requirements compared to autodiff tools.
We present a method for efficient differentiable simulation of articulated bodies. This enables integration of articulated body dynamics into deep learning frameworks, and gradient-based optimization of neural networks that operate on articulated bodies. We derive the gradients of the forward dynamics using spatial algebra and the adjoint method. Our approach is an order of magnitude faster than autodiff tools. By only saving the initial states throughout the simulation process, our method reduces memory requirements by two orders of magnitude. We demonstrate the utility of efficient differentiable dynamics for articulated bodies in a variety of applications. We show that reinforcement learning with articulated systems can be accelerated using gradients provided by our method. In applications to control and inverse problems, gradient-based optimization enabled by our work accelerates convergence by more than an order of magnitude.