Data-driven Policy Transfer with Imprecise Perception Simulation
This addresses robot motion control in real-world environments with imprecise simulation, but it appears incremental as it builds on existing simulation and learning methods.
The paper tackles learning continuous motion control policies for a mobile robot using a non-differentiable physics simulator and complex state estimation, achieving this through a coarse-to-fine learning paradigm with joint optimization of the policy and a generative model, and evaluates it on a real-world platform.
The paper presents a complete pipeline for learning continuous motion control policies for a mobile robot when only a non-differentiable physics simulator of robot-terrain interactions is available. The multi-modal state estimation of the robot is also complex and difficult to simulate, so we simultaneously learn a generative model which refines simulator outputs. We propose a coarse-to-fine learning paradigm, where the coarse motion planning is alternated with imitation learning and policy transfer to the real robot. The policy is jointly optimized with the generative model. We evaluate the method on a real-world platform in a batch of experiments.