SO-LOST - An Ant-Trail Algorithm for Multi-Robot Navigation with Active Interference Reduction.

Abstract

We consider a group of autonomous robots which perform the classical task of transporting resources from a source to home. The robots use ant-like emergent trail following to navigate between home and source. When trails lie close together, spatial interference between robots navigating in opposite directions reduces overall system performance. This paper proposes a navigation strategy which is effective in separating trails with different goals. The results of simulation experiments indicate that the performance of robots is usefully increased compared to original algorithm in constrained environments.