A Survivability Enhanced Swarm Robotic Searching System Using Multi-objective Particle Swarm Optimization

摘要

This paper aims at outlining an algorithm for groups of swarm robots solely powered by light energy to survive and complete target searching tasks in unknown fields where light energy charging points and targets are scattered. To sustain the searching operation and solve energy consumption conflicts between surviving and searching, this paper introduces a multi-robot algorithm based on Multi-Objective Particle Swarm Optimization (MOPSO) and energy-saving decision rules. A novel mechanism of selecting the best performing particle in PSO is introduced. Several sets of simulation experiments were conducted and results show that a 15-robot swarm system running this algorithm is able to search a single target and stabilize the energy level for the long-term simultaneously. It demonstrates the feasibility of applying this energy-optimized MOPSO as a design framework for a long-term searching swarm robot system.