In-Hand Re-grasp Manipulation with Passive Dynamic Actions via Imitation Learning

TL;DR

This paper introduces an imitation learning-based in-hand re-grasp controller that uses minimal object information, achieving high success rates in diverse physical tasks with varying object properties.

Contribution

It presents a novel end-to-end sliding motion controller trained via GAIL that adapts to different objects without detailed contact or geometry data.

Findings

Achieved 86% success rate in physical experiments.

Outperformed baseline algorithms like BC and PPO.

Demonstrated versatility across various object types.

Abstract

Re-grasp manipulation leverages on ergonomic tools to assist humans in accomplishing diverse tasks. In certain scenarios, humans often employ external forces to effortlessly and precisely re-grasp tools like a hammer. Previous development on controllers for in-grasp sliding motion using passive dynamic actions (e.g.,gravity) relies on apprehension of finger-object contact information, and requires customized design for individual objects with varied geometry and weight distribution. It limits their adaptability to diverse objects. In this paper, we propose an end-to-end sliding motion controller based on imitation learning (IL) that necessitates minimal prior knowledge of object mechanics, relying solely on object position information. To expedite training convergence, we utilize a data glove to collect expert data trajectories and train the policy through Generative Adversarial Imitation Learning (GAIL). Simulation results demonstrate the controller's versatility in performing in-hand sliding tasks with objects of varying friction coefficients, geometric shapes, and masses. By migrating to a physical system using visual position estimation, the controller demonstrated an average success rate of 86%, surpassing the baseline algorithm's success rate of 35% of Behavior Cloning(BC) and 20% of Proximal Policy Optimization (PPO).