Raising Body Ownership in End-to-End Visuomotor Policy Learning via Robot-Centric Pooling

TL;DR

This paper introduces Robot-centric Pooling (RcP), a novel method that improves visuomotor policy robustness by emphasizing robot-related image features, trained jointly with policies without extra data, and tested on reaching tasks in simulation and real-world.

Contribution

The paper presents RcP, a new pooling technique that enhances end-to-end visuomotor policies by focusing on robot-centric features, improving robustness against distractors and pixel shifts.

Findings

RcP significantly improves policy robustness to unseen distractors.

RcP enhances resilience to pixel shifts compared to baseline methods.

The method is effective in both simulated and real-world reaching tasks.

Abstract

We present Robot-centric Pooling (RcP), a novel pooling method designed to enhance end-to-end visuomotor policies by enabling differentiation between the robots and similar entities or their surroundings. Given an image-proprioception pair, RcP guides the aggregation of image features by highlighting image regions correlating with the robot's proprioceptive states, thereby extracting robot-centric image representations for policy learning. Leveraging contrastive learning techniques, RcP integrates seamlessly with existing visuomotor policy learning frameworks and is trained jointly with the policy using the same dataset, requiring no extra data collection involving self-distractors. We evaluate the proposed method with reaching tasks in both simulated and real-world settings. The results demonstrate that RcP significantly enhances the policies' robustness against various unseen distractors, including self-distractors, positioned at different locations. Additionally, the inherent robot-centric characteristic of RcP enables the learnt policy to be far more resilient to aggressive pixel shifts compared to the baselines.