Augmenting Physical Simulators with Stochastic Neural Networks: Case Study of Planar Pushing and Bouncing
This addresses the need for accurate and fast simulators in robotics for state estimation, planning, and control, representing an incremental improvement by combining existing methods.
The paper tackled the problem of creating an efficient, generalizable physical simulator with uncertainty estimates for robot applications by augmenting an analytical rigid-body simulator with a neural network to model residuals, achieving consistent outperformance over purely analytical and learned simulators on real benchmarks.
An efficient, generalizable physical simulator with universal uncertainty estimates has wide applications in robot state estimation, planning, and control. In this paper, we build such a simulator for two scenarios, planar pushing and ball bouncing, by augmenting an analytical rigid-body simulator with a neural network that learns to model uncertainty as residuals. Combining symbolic, deterministic simulators with learnable, stochastic neural nets provides us with expressiveness, efficiency, and generalizability simultaneously. Our model outperforms both purely analytical and purely learned simulators consistently on real, standard benchmarks. Compared with methods that model uncertainty using Gaussian processes, our model runs much faster, generalizes better to new object shapes, and is able to characterize the complex distribution of object trajectories.