Follow the Gradient: Crossing the Reality Gap using Differentiable Physics (RealityGrad)
This addresses the problem of transferring robot control from simulation to reality efficiently, with incremental improvements in method and application.
The paper tackles the reality gap in robotics by proposing RealityGrad, an iterative method that uses differentiable physics for sim2real transfer and real2sim model optimization, reducing error by two-thirds in under 22 minutes per iteration on a desktop computer.
We propose a novel iterative approach for crossing the reality gap that utilises live robot rollouts and differentiable physics. Our method, RealityGrad, demonstrates for the first time, an efficient sim2real transfer in combination with a real2sim model optimisation for closing the reality gap. Differentiable physics has become an alluring alternative to classical rigid-body simulation due to the current culmination of automatic differentiation libraries, compute and non-linear optimisation libraries. Our method builds on this progress and employs differentiable physics for efficient trajectory optimisation. We demonstrate RealitGrad on a dynamic control task for a serial link robot manipulator and present results that show its efficiency and ability to quickly improve not just the robot's performance in real world tasks but also enhance the simulation model for future tasks. One iteration of RealityGrad takes less than 22 minutes on a desktop computer while reducing the error by 2/3, making it efficient compared to other sim2real methods in both compute and time. Our methodology and application of differentiable physics establishes a promising approach for crossing the reality gap and has great potential for scaling to complex environments.