Contact Coverage-Guided Exploration for General-Purpose Dexterous Manipulation

TL;DR

This paper introduces Contact Coverage-Guided Exploration (CCGE), a novel exploration method for general-purpose dexterous manipulation that encourages discovering diverse contact patterns, significantly improving training efficiency and success rates in complex tasks.

Contribution

CCGE is the first exploration approach to utilize contact coverage and energy-based rewards for general-purpose dexterous manipulation tasks.

Findings

CCGE improves training efficiency and success rates over existing methods.

Contact patterns learned with CCGE transfer effectively to real-world robots.

CCGE enables discovery of diverse and novel contact interactions.

Abstract

Deep Reinforcement learning (DRL) has achieved remarkable success in domains with well-defined reward structures, such as Atari games and locomotion. In contrast, dexterous manipulation lacks general-purpose reward formulations and typically depends on task-specific, handcrafted priors to guide hand-object interactions. We propose Contact Coverage-Guided Exploration (CCGE), a general exploration method designed for general-purpose dexterous manipulation tasks. CCGE represents contact state as the intersection between object surface points and predefined hand keypoints, encouraging dexterous hands to discover diverse and novel contact patterns, namely which fingers contact which object regions. It maintains a contact counter conditioned on discretized object states obtained via learned hash codes, capturing how frequently each finger interacts with different object regions. This counter is leveraged in two complementary ways: (1) to assign a count-based contact coverage reward that promotes exploration of novel contact patterns, and (2) an energy-based reaching reward that guides the agent toward under-explored contact regions. We evaluate CCGE on a diverse set of dexterous manipulation tasks, including cluttered object singulation, constrained object retrieval, in-hand reorientation, and bimanual manipulation. Experimental results show that CCGE substantially improves training efficiency and success rates over existing exploration methods, and that the contact patterns learned with CCGE transfer robustly to real-world robotic systems. Project page is https://contact-coverage-guided-exploration.github.io.