Efficient Sim-to-real Transfer of Contact-Rich Manipulation Skills with Online Admittance Residual Learning

TL;DR

This paper presents a hybrid offline-online reinforcement learning framework that improves the transfer of contact-rich manipulation skills from simulation to real robots by learning residual compliance control parameters in real time.

Contribution

It introduces a novel approach combining offline simulation-based learning with online residual adaptation for robust sim-to-real transfer in contact-rich tasks.

Findings

Outperforms existing methods in assembly, pivoting, and screwing tasks.

Enhances robustness and safety in contact-rich manipulation.

Reduces sim-to-real gap through online residual learning.

Abstract

Learning contact-rich manipulation skills is essential. Such skills require the robots to interact with the environment with feasible manipulation trajectories and suitable compliance control parameters to enable safe and stable contact. However, learning these skills is challenging due to data inefficiency in the real world and the sim-to-real gap in simulation. In this paper, we introduce a hybrid offline-online framework to learn robust manipulation skills. We employ model-free reinforcement learning for the offline phase to obtain the robot motion and compliance control parameters in simulation \RV{with domain randomization}. Subsequently, in the online phase, we learn the residual of the compliance control parameters to maximize robot performance-related criteria with force sensor measurements in real time. To demonstrate the effectiveness and robustness of our approach, we provide comparative results against existing methods for assembly, pivoting, and screwing tasks.