DexArt: Benchmarking Generalizable Dexterous Manipulation with Articulated Objects

TL;DR

This paper introduces DexArt, a benchmark for evaluating generalizable dexterous manipulation of diverse articulated objects using reinforcement learning and 3D representation learning in simulation.

Contribution

The paper presents DexArt, a new benchmark for dexterous manipulation with articulated objects, and investigates the impact of 3D representation learning on policy generalization.

Findings

3D representation learning improves decision making in RL for manipulation tasks.

DexArt enables evaluation of policies on unseen articulated objects.

Insights into how 3D inputs influence manipulation success.

Abstract

To enable general-purpose robots, we will require the robot to operate daily articulated objects as humans do. Current robot manipulation has heavily relied on using a parallel gripper, which restricts the robot to a limited set of objects. On the other hand, operating with a multi-finger robot hand will allow better approximation to human behavior and enable the robot to operate on diverse articulated objects. To this end, we propose a new benchmark called DexArt, which involves Dexterous manipulation with Articulated objects in a physical simulator. In our benchmark, we define multiple complex manipulation tasks, and the robot hand will need to manipulate diverse articulated objects within each task. Our main focus is to evaluate the generalizability of the learned policy on unseen articulated objects. This is very challenging given the high degrees of freedom of both hands and objects. We use Reinforcement Learning with 3D representation learning to achieve generalization. Through extensive studies, we provide new insights into how 3D representation learning affects decision making in RL with 3D point cloud inputs. More details can be found at https://www.chenbao.tech/dexart/.