Force-Aware Residual DAgger via Trajectory Editing for Precision Insertion with Impedance Control

2026-03-04

AI Summary

This work addresses the deployment challenges of imitation learning in precision insertion tasks, which are exacerbated by covariate shift and frequent expert interventions. To overcome these issues, the authors propose the TER-DAgger framework, which learns a residual policy through optimization-based trajectory editing that seamlessly integrates human corrections with autonomous execution. A key innovation is the incorporation of a force-aware failure prediction mechanism within a Cartesian impedance control framework, enabling stable and safe contact-rich manipulation. By synergistically combining trajectory editing, force-triggered feedback, and impedance control, TER-DAgger substantially mitigates covariate shift and reduces the need for expert intervention. Experimental results in both simulation and real-world settings demonstrate that TER-DAagger achieves an average success rate over 37% higher than baseline approaches—including behavioral cloning, manual correction, and retraining—on precision insertion tasks.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Intelligent Robots: Manipulation
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Imitation learning (IL) has shown strong potential for contact-rich precision insertion tasks. However, its practical deployment is often hindered by covariate shift and the need for continuous expert monitoring to recover from failures during execution. In this paper, we propose Trajectory Editing Residual Dataset Aggregation (TER-DAgger), a scalable and force-aware human-in-the-loop imitation learning framework that mitigates covariate shift by learning residual policies through optimization-based trajectory editing. This approach smoothly fuses policy rollouts with human corrective trajectories, providing consistent and stable supervision. Second, we introduce a force-aware failure anticipation mechanism that triggers human intervention only when discrepancies arise between predicted and measured end-effector forces, significantly reducing the requirement for continuous expert monitoring. Third, all learned policies are executed within a Cartesian impedance control framework, ensuring compliant and safe behavior during contact-rich interactions. Extensive experiments in both simulation and real-world precision insertion tasks show that TER-DAgger improves the average success rate by over 37\% compared to behavior cloning, human-guided correction, retraining, and fine-tuning baselines, demonstrating its effectiveness in mitigating covariate shift and enabling scalable deployment in contact-rich manipulation.

Problem

• covariate shift
• imitation learning
• precision insertion
• expert monitoring
• contact-rich manipulation

Innovation

• Imitation Learning
• Residual Policy
• Trajectory Editing
• Force-Aware Failure Anticipation
• Impedance Control