Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections

Xiaomeng Xu, Yifan Hou, Zeyi Liu, Shuran Song

Keywords: Robotic manipulation, Imitation learning, Data Aggregation

TL;DR: We propose a system to collect high quality human correction data on a pretrained policy and a method to learn correction behavior from the data.

Abstract:

We address key challenges in Dataset Aggregation (DAgger) for real-world contact-rich manipulation: how to collect informative human correction data and how to effectively update policies with this new data. We introduce Compliant Residual DAgger (CR-DAgger), which contains two novel components:
1) a Compliant Intervention Interface that leverages compliance control, allowing humans to provide gentle, accurate delta action corrections without interrupting the ongoing robot policy execution;
2) a Compliant Residual Policy formulation that learns from human corrections while incorporating force feedback and force control.

Our system significantly enhances performance on precise contact-rich manipulation tasks using minimal correction data, improving base policy success rates by over 50\% on two challenging tasks (book flipping and belt assembly) while outperforming both retraining-from-scratch and finetuning approaches. Through extensive real-world experiments, we provide practical guidance for implementing effective DAgger in real-world robot learning tasks.

Submission Number: 16