Decentralised self-organising maps for the online orienteering problem with neighbourhoods

摘要

This extended abstract presents a new coordination algorithm for decentralised multi-robot information gathering. We consider planning for an online variant of the multi-agent orienteering problem with neighbourhoods. This formulation closely aligns with a number of important tasks in robotics, including inspection, surveillance, and reconnaissance. We propose a decentralised variant of the self-organising map (SOM) learning procedure, named Dec-SOM, which efficiently finds plans for a team of robots in a non-myopic manner. Decentralisation is achieved by performing a distributed allocation scheme jointly with the SOM adaptations. Preliminary simulation results indicate that Dec-SOM outperforms baseline methods and is a viable solution for decentralised, online, and non-myopic information gathering.