Dataset: Purely vision-based collective movement of robots

Collective movement inspired by animal groups promises inherited benefits for robot swarms, such as enhanced sensing and efficiency. However, while animals move in groups using only their local senses, robots often obey central control or use direct communication, introducing systemic weaknesses to the swarm. In the hope of addressing such vulnerabilities, developing bio-inspired decentralized swarms has been a major focus in recent decades. Yet, creating robots that move efficiently together using only local sensory information remains an extraordinary challenge. In this work, we present a decentralized, purely vision-based swarm of ten terrestrial robots, the first of its kind. Within our framework, robots achieve collisionless, polarized motion through minimal visual interactions, computing everything on board based exclusively on their camera streams, making central processing or direct communication obsolete. With agent-based simulations, we further show that using this model, even with a strictly limited field of view and within confined spaces, ordered group motion can emerge, while also highlighting key limitations. Our results set the stage for advanced, decentralized vision-based robotic swarms capable of diverse tasks and offer a multitude of practical applications.

受动物群体行为启发的集群运动，可为机器人群（robot swarm）带来诸如感知能力增强、运行效率提升等固有优势。然而，动物群体移动仅依赖局部感知，而现有机器人群往往采用集中式控制或直接通信机制，这会为集群引入系统性脆弱点。为解决此类缺陷，近几十年来，开发受生物启发的分布式机器人群已成为研究热点，但仅依靠局部感知信息实现高效协同移动的机器人群仍是一项极具挑战性的难题。本研究提出了同类首个纯视觉驱动的分布式地面机器人群，共包含10台地面机器人；在该研究框架下，机器人仅通过摄像头数据流在本地板载设备完成所有计算，无需依赖集中处理或直接通信，仅依靠极简的视觉交互即可实现无碰撞的极化集群运动。借助基于智能体（agent）的仿真实验，我们进一步验证了该模型的有效性：即便在视场严格受限且空间封闭的环境中，有序的群体运动仍可自发涌现，同时我们也点明了该方法的核心局限性。本研究成果为可承担多样化任务的先进分布式视觉驱动机器人群的研发奠定了技术基础，并为诸多实际应用场景提供了可行路径。