Adaptation of the Ant Algorithm to Control a Robot Swarm

Abstract

In the work the prospects for using the ant algorithm to control a robot swarm is studied. It has been shown that the most important features of the ant algorithm for controlling a robot swarm are: lack of centralized management; the ability of members of the swarm to be guided exclusively by local information; the efficiency of the swarm, even with the loss of a large number of its members. It was proposed to adapt the ant algorithm by using the reference map and individual map of pheromone trail stored in the memory of each robot. The efficiency of the proposed algorithm was proved.