Collective Ranking of Environmental Signals through Gaussian Belief Propagation in a Patrolling Robot Swarm
Date:
This was my presentation of the paper accepted at DARS 2026, co-authored with Connor York, Jonathan Lawry and Edmund R. Hunt. The talk covered how the patrol graph that dictates where a robot swarm moves can also serve as a factor graph, allowing Gaussian Belief Propagation to form a collective ranking of all patrol locations by a measured signal. I compared this against simple and visit-weighted averaging in simulation, and showed hardware trials with four Leo Rovers tracking a radio signal in an office lobby. The paper is available here.