Admittance Visuomotor Policy Learning for General-Purpose Contact-Rich Manipulations

TL;DR

This paper introduces an admittance visuomotor policy framework for contact-rich manipulation tasks, leveraging multimodal data and diffusion models to improve success rates, efficiency, and compliance in robotic manipulation.

Contribution

It presents a novel framework combining diffusion-based planning and admittance control for general-purpose contact-rich manipulation, with a user-friendly teleoperation system for data collection.

Findings

Achieved higher success rates than state-of-the-art methods.

Reduced contact force by an average of 48.8%.

Demonstrated smoother, more efficient contact handling.

Abstract

Contact force in contact-rich environments is an essential modality for robots to perform general-purpose manipulation tasks, as it provides information to compensate for the deficiencies of visual and proprioceptive data in collision perception, high-precision grasping, and efficient manipulation. In this paper, we propose an admittance visuomotor policy framework for continuous, general-purpose, contact-rich manipulations. During demonstrations, we designed a low-cost, user-friendly teleoperation system with contact interaction, aiming to gather compliant robot demonstrations and accelerate the data collection process. During training and inference, we propose a diffusion-based model to plan action trajectories and desired contact forces from multimodal observation that includes contact force, vision and proprioception. We utilize an admittance controller for compliance action execution. A comparative evaluation with two state-of-the-art methods was conducted on five challenging tasks, each focusing on different action primitives, to demonstrate our framework's generalization capabilities. Results show our framework achieves the highest success rate and exhibits smoother and more efficient contact compared to other methods, the contact force required to complete each tasks was reduced on average by 48.8%, and the success rate was increased on average by 15.3%. Videos are available at https://ryanjiao.github.io/AdmitDiffPolicy/.