Multi-Stage Reinforcement Learning for Non-Prehensile Manipulation

TL;DR

This paper introduces MRLM, a multi-stage reinforcement learning framework for non-prehensile object manipulation, enabling flexible, occlusion-robust, and zero-shot transferable manipulation skills with significant success rate improvements.

Contribution

The paper proposes a novel multi-stage RL approach with new representations and metrics, advancing non-prehensile manipulation capabilities beyond previous single-skill methods.

Findings

Success rate improved by 40-100% over baselines.

Achieves 95% success rate in real-world zero-shot transfer.

Demonstrates strong generalization to unseen object shapes.

Abstract

Manipulating objects without grasping them enables more complex tasks, known as non-prehensile manipulation. Most previous methods only learn one manipulation skill, such as reach or push, and cannot achieve flexible object manipulation.In this work, we introduce MRLM, a Multi-stage Reinforcement Learning approach for non-prehensile Manipulation of objects.MRLM divides the task into multiple stages according to the switching of object poses and contact points.At each stage, the policy takes the point cloud-based state-goal fusion representation as input, and proposes a spatially-continuous action that including the motion of the parallel gripper pose and opening width.To fully unlock the potential of MRLM, we propose a set of technical contributions including the state-goal fusion representation, spatially-reachable distance metric, and automatic buffer compaction.We evaluate MRLM on an Occluded Grasping task which aims to grasp the object in configurations that are initially occluded.Compared with the baselines, the proposed technical contributions improve the success rate by at least 40\% and maximum 100\%, and avoids falling into local optimum.Our method demonstrates strong generalization to unseen object with shapes outside the training distribution.Moreover, MRLM can be transferred to real world with zero-shot transfer, achieving a 95\% success rate.Code and videos can be found at https://sites.google.com/view/mrlm.