Optimal PSO for collective robotic search applications

Unmanned vehicles/mobile robots are of particular interest in target tracing applications since there are many areas where a human cannot explore. Different means of control have been investigated for unmanned vehicles with various algorithms like genetic algorithms, evolutionary computations, neural networks etc. This work presents the application of particle swarm optimization (PSO) for collective robotic search. The performance of the PSO algorithm depends on various parameters called quality factors and these parameters are determined using a secondary PSO. Results are presented to show that the performance of PSO algorithm and search is improved for a single and multiple target searches.