Master Micro Residual Correction with Adaptive Tactile Fusion and Force-Mixed Control for Contact-Rich Manipulation

2026-03-16

AI Summary

This work addresses the perception-action challenges in contact-intensive fine manipulation tasks, where complex interaction dynamics and multi-timescale control conflicts hinder effective execution. To this end, the authors propose the M2-ResiPolicy architecture, which integrates a low-frequency diffusion-based main policy with a high-frequency lightweight GRU micro-corrector. This framework further incorporates a tactile-intensity-driven adaptive visual-tactile fusion mechanism and a force-mixed proportional-based impedance controller (PBIC), enabling unified global task planning and millisecond-level local responsiveness. Evaluated on a delicate chip grasping task, the method achieves a 93% damage-free success rate, significantly outperforming baseline approaches such as Diffusion Policy, thereby demonstrating its superior capability in high-precision force regulation and stable contact control.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Interaction Techniques and Devices
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Robotic contact-rich and fine-grained manipulation remains a significant challenge due to complex interaction dynamics and the competing requirements of multi-timescale control. While current visual imitation learning methods excel at long-horizon planning, they often fail to perceive critical interaction cues like friction variations or incipient slip, and struggle to balance global task coherence with local reactive feedback. To address these challenges, we propose M2-ResiPolicy, a novel Master-Micro residual control architecture that synergizes high-level action guidance with low-level correction. The framework consists of a Master-Guidance Policy (MGP) operating at 10 Hz, which generates temporally consistent action chunks via a diffusion-based backbone and employs a tactile-intensity-driven adaptive fusion mechanism to dynamically modulate perceptual weights between vision and touch. Simultaneously, a high-frequency (60 Hz) Micro-Residual Corrector (MRC) utilizes a lightweight GRU to provide real-time action compensation based on TCP wrench feedback. This policy is further integrated with a force-mixed PBIC execution layer, effectively regulating contact forces to ensure interaction safety. Experiments across several demanding tasks including fragile object grasping and precision insertion, demonstrate that M2-ResiPolicy significantly outperforms standard Diffusion Policy (DP) and state-of-the-art Reactive Diffusion Policy (RDP), achieving a 93\% damage-free success rate in chip grasping and superior force regulation stability.

Problem

• contact-rich manipulation
• fine-grained manipulation
• multi-timescale control
• tactile perception
• interaction dynamics

Innovation

• residual control
• adaptive tactile fusion
• force-mixed control
• diffusion policy
• contact-rich manipulation