Deformable Cluster Manipulation via Whole-Arm Policy Learning

2025-07-22

AI Summary

Deformable object swarm manipulation faces challenges including weak model generalization, high perception uncertainty, and low spatial abstraction efficiency. Method: We propose a whole-arm contact-aware reinforcement learning framework that fuses 3D point clouds with proprioceptive tactile signals to construct a distributional state representation; kernel mean embedding enables high-dimensional uncertain state modeling. An occlusion-aware, context-agnostic heuristic strategy supports zero-shot cross-scenario transfer. Departing from end-effector-centric paradigms, we leverage multi-link coordination for active occlusion removal. Results: Evaluated on power-line obstacle clearance, the system generates novel, adaptive policies without real-data fine-tuning—successfully handling unknown occlusions and dynamic deformations. It demonstrates significant advantages in sim-to-real transfer, dexterous manipulation generalization, and robust physical interaction.

Technology Category

• Intelligent Robots: Manipulation
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work

Abstract

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.

Problem

• Manipulating deformable object clusters with whole-arm interactions
• Addressing limited model synthesis and high perception uncertainty
• Developing efficient spatial abstractions for deformable object manipulation

Innovation

• Model-free policy learning with 3D point clouds
• Kernel mean embeddings for efficient training
• Context-agnostic occlusion heuristic for deformables