Excavation Reinforcement Learning Using Geometric Representation

Excavation of irregular rigid objects in clutter, such as fragmented rocks and wood blocks, is very challenging due to their complex interaction dynamics and highly variable geometries. In this paper, we adopt reinforcement learning (RL) to tackle this challenge and learn policies to plan for a sequence of excavation trajectories for irregular rigid objects, given point clouds of excavation scenes. Moreover, we separately learn a compact representation of the point cloud on geometric tasks that do not require human labeling. We show that using the representation reduces training time for RL, while achieving similar asymptotic performance compare to an end-to-end RL algorithm. Our simulated and physical experiments also demonstrate that the learned representation improves the generalization of the learned RL policy. To our best knowledge, this paper presents the first application of RL to excavation of irregular rigid objects in clutter.