Rethinking Optimization with Differentiable Simulation from a Global Perspective
This work addresses optimization difficulties in robotics and simulation for scenarios involving deformable objects and fluids, representing an incremental improvement over existing differentiable simulation methods.
The paper tackles the challenge of optimizing policies in contact-rich scenarios with differentiable simulation, where gradients can be rugged, by proposing a method combining Bayesian optimization with semi-local leaps, resulting in outperformance over baselines in simulation and real-robot experiments.
Differentiable simulation is a promising toolkit for fast gradient-based policy optimization and system identification. However, existing approaches to differentiable simulation have largely tackled scenarios where obtaining smooth gradients has been relatively easy, such as systems with mostly smooth dynamics. In this work, we study the challenges that differentiable simulation presents when it is not feasible to expect that a single descent reaches a global optimum, which is often a problem in contact-rich scenarios. We analyze the optimization landscapes of diverse scenarios that contain both rigid bodies and deformable objects. In dynamic environments with highly deformable objects and fluids, differentiable simulators produce rugged landscapes with nonetheless useful gradients in some parts of the space. We propose a method that combines Bayesian optimization with semi-local 'leaps' to obtain a global search method that can use gradients effectively, while also maintaining robust performance in regions with noisy gradients. We show that our approach outperforms several gradient-based and gradient-free baselines on an extensive set of experiments in simulation, and also validate the method using experiments with a real robot and deformables. Videos and supplementary materials are available at https://tinyurl.com/globdiff