Low-Communication Task Coordination Strategy for Multi-Robot Collaboration

Abstract

Multi-robot task coordination often relies on continuous communication to maintain consistent task allocation and avoid conflicts. However, in real-world scenarios with limited network bandwidth, periodic broadcasting may introduce substantial communication overhead and increase the risk of channel congestion. This paper proposes a lightweight, low-communication task coordination strategy for decentralised multi-robot systems. The method combines an event-triggered communication mechanism with a lexicographical distributed arbitration rule. Robots communicate only when task-status updates or proximity events occur. The conflicting task selections are then exposed and resolved reactively through distributed arbitration. Extensive simulations under different team scales and task densities show that the proposed strategy matches the temporal efficiency of full-communication approaches while reducing network load. It cuts communication overhead by up to 85% in dense-task simulations and improves coordination efficiency in dense-task scenarios. These results confirm that efficient coordination is possible with a minimal communication footprint, laying the groundwork for future physical validation.