Deformable Cluster Manipulation via Whole-Arm Policy Learning
This addresses the challenge of contact-rich manipulation of deformable objects in real-world applications like infrastructure maintenance, representing an incremental advance in robotic manipulation.
The paper tackles the problem of manipulating clusters of deformable objects, such as clearing branches in power line maintenance, by developing a model-free reinforcement learning framework that integrates 3D point clouds and touch indicators to enable whole-arm interactions. The result includes successful zero-shot sim-to-real policy transfer, allowing the arm to clear real branches with unknown occlusion patterns and uncertain dynamics.
Manipulating clusters of deformable objects presents a substantial challenge with widespread applicability, but requires contact-rich whole-arm interactions. A potential solution must address the limited capacity for realistic model synthesis, high uncertainty in perception, and the lack of efficient spatial abstractions, among others. We propose a novel framework for learning model-free policies integrating two modalities: 3D point clouds and proprioceptive touch indicators, emphasising manipulation with full body contact awareness, going beyond traditional end-effector modes. Our reinforcement learning framework leverages a distributional state representation, aided by kernel mean embeddings, to achieve improved training efficiency and real-time inference. Furthermore, we propose a novel context-agnostic occlusion heuristic to clear deformables from a target region for exposure tasks. We deploy the framework in a power line clearance scenario and observe that the agent generates creative strategies leveraging multiple arm links for de-occlusion. Finally, we perform zero-shot sim-to-real policy transfer, allowing the arm to clear real branches with unknown occlusion patterns, unseen topology, and uncertain dynamics.