Intelligent control technique for autonomous collective robotics systems

Abstract

The paper considers the problem of finding maximum number of the radiation sources that are located on the plain field by group of mobile agents with limited time. We formulate both physical and mathematical models that describe the environment and the behavior of mobile agents by including a topology of communication between agents. We propose MPSO-Optima which is improved version of Particle Swarm Optimization (PSO) algorithm adapted to changes in the velocity directions by the conjugate gradient method and optimizing magnitude of the inertia ratio. Both improvements lead to the higher convergence rate that proved by numerical experiments. As a result of the study, the method demonstrates increased resistance to the variation of the acceleration coefficients. The algorithm is tested on virtual simulator environment.