H. Hickson, S. Lee, S. Jones, T. Didiot-Cook, S. Hauert and A. Mavromatis, "Less is More: An Analysis of Minimal Information Sharing on the Performance of Robot Swarms," 2025 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C), Tokyo, Japan, 2025, pp. 1-8, doi: 10.1109/ACSOS-C66519.2025.00024.
Abstract: This paper investigates the impact of minimal global information sharing on the coordination and performance of robot swarms. We focus on lightweight task-related data and assess the impact of different information sharing strategies on task performance. Through both simulations and real-world experiments, we evaluate performance under varying swarm sizes, task complexities, and communication architectures. Results show that minimal global information can significantly enhance performance in small to mid-sized swarms, but may hinder coordination when not carefully selected for a given scenario or swarm size. Additionally, we demonstrate that real-world communication constraints, including latency and message-passing limitations, do not significantly degrade performance, indicating the practicality of minimal information sharing in the real world. This study highlights the importance of carefully selecting and aligning shared information with task characteristics.
More Information
