Evaluation of a Coordination Model for Robot Swarms in Collective Environment Mapping

Collective environmental mapping leverages the decentralised and cooperative principles of swarm robotics to efficiently explore and map complex environments. Swarm robotics is particularly advantageous in large-scale or dynamic scenarios, where centralised systems often face limitations related to coordination overhead and computational scalability. This work enhances a bio-inspired coordination model for robot swarms by integrating a queue-based Past-Path Memory. While the original model enables effective swarm distribution and periodic environment revisitation, the memory-enhanced adaptation allows robots to retain and share environmental data, promoting the emergence of a global mapping behaviour. Experimental results demonstrate that this adaptation significantly improves collective mapping efficiency, with performance gains ranging from 280% to 857% over the baseline model and scaling proportionally with Past-Path Memory size.