Swarm Optimization Approach for Real Light Source Detection of Multi-robot System

Abstract

Exploration and searching in unknown or hazardous environments using Multirobot System (MRS) is among the principal topics in robotics. There have been numerous researches on searching and detection of odor, fire or pollution sources. In this paper, we present results on detecting real light sources with MRS using a modified Particle Swarm Optimization Algorithm (PSO), namely APSO. In the modified algorithm, an integration of conventional PSO and Artificial Potential Field (APF) is employed to translate the idea of using swarm intelligence for space exploration and light source detection into reality. The formula for PSO component velocities are devised with the introduction of APF. Furthermore, each particle is surrounded by an APF that forms repulsive forces to prevent collision while the swarm is in operation. Any particle would find itself within a time-varying potential field, as the distribution of its surrounding particles as well as global and local best positions continually change when the exploration makes progress. The simulation results of APSO on Matlab environment in various scenarios show the reliability and efficiency of the modified algorithm.