Trophallaxis and energy optimization in swarms of robots

Abstract

Ability to allocate task on the fly is considered to be one of the most desirable features in a swarm intelligent system. This paper presents a computational model in which swarms of autonomous agents (robots) carry out the task of cleaning the environment by collecting boxes from the environment and dumping them in a dump area. As agents work, they lose energy and when the energy is too low they need to go to the charging area to gain energy. Our model is inspired by how social insects and in particular how ants behave. Experimental results show that incorporating trophallactic behavior in swarms of robots improve the performance of the swarm in terms of the energy consumption over earlier strategies. The proposed model is found to be efficient, accurate and consistent with the biological equivalents.