Learning Coordinated Visuomotor Box-Pushing from Solo Demonstrations

2026-10-05

AI Summary

This study addresses the challenge of collecting collective demonstration data in multi-robot collaboration by proposing a novel paradigm that achieves large-scale, communication-free coordination using only single-robot teleoperation data. By training visuomotor policies combined with decentralized onboard control, the approach enables implicit coordination in box-pushing tasks while minimizing interference. Furthermore, this work systematically evaluates the effectiveness of dataset construction strategies and lightweight architectures. Experiments involving up to 40 robots validate the feasibility of the proposed method and reveal a core bottleneck: relying solely on passive observation of other agents' behaviors is insufficient for acquiring effective coordination. These findings provide critical insights into learning multi-agent coordination from single-agent demonstrations.

Abstract

Multi-robot imitation learning, particularly in settings where visuomotor policies are deployed in a communication-free, onboard decentralised style, represents an attractive paradigm. However, its realisation remains insufficiently understood, largely due to the difficulty of collecting collective demonstrations, since a single operator cannot control many robots simultaneously. Meanwhile, unlike coupled collaborative manipulation, many coordinated tasks achieve system-wide efficiency primarily through minimising inter-robot interference. This structure motivates us to study whether data collected by a teleoperated single-robot can be leveraged for large-scale coordinated box-pushing as a testbed. We systematically investigate dataset creation strategies and lightweight policy architectures. In particular, experiments with up to 40 robots highlight the difficulty of acquiring effective coordination solely through passive observation of other operating robots, revealing a concrete bottleneck for multi-robot research.

Problem

• multi-robot imitation learning
• visuomotor policies
• coordinated box-pushing
• solo demonstrations
• decentralised control

Innovation

• Multi-robot imitation learning
• Visuomotor policies
• Decentralised coordination
• Solo demonstrations
• Box-pushing