Enhancing Tactile-based Reinforcement Learning for Robotic Control

TL;DR

This paper improves tactile-based reinforcement learning for robotic manipulation by developing self-supervised methods, demonstrating superhuman dexterity in contact tasks, and introducing the RoTO benchmark to advance research in tactile sensing.

Contribution

It introduces scalable SSL techniques for tactile RL, highlights the importance of sparse tactile signals, and releases a new benchmark for future research.

Findings

Sparse tactile signals are crucial for dexterity.

Decoupling SSL memory from on-policy memory improves performance.

Agents outperform humans in complex contact tasks.

Abstract

Achieving safe, reliable real-world robotic manipulation requires agents to evolve beyond vision and incorporate tactile sensing to overcome sensory deficits and reliance on idealised state information. Despite its potential, the efficacy of tactile sensing in reinforcement learning (RL) remains inconsistent. We address this by developing self-supervised learning (SSL) methodologies to more effectively harness tactile observations, focusing on a scalable setup of proprioception and sparse binary contacts. We empirically demonstrate that sparse binary tactile signals are critical for dexterity, particularly for interactions that proprioceptive control errors do not register, such as decoupled robot-object motions. Our agents achieve superhuman dexterity in complex contact tasks (ball bouncing and Baoding ball rotation). Furthermore, we find that decoupling the SSL memory from the on-policy memory can improve performance. We release the Robot Tactile Olympiad (RoTO) benchmark to standardise and promote future research in tactile-based manipulation. Project page: https://elle-miller.github.io/tactile_rl