Statistical Significance Analysis of the R-DPSO (M. S. Couceiro, N. Ferreira, R. P. Rocha, F. M. L. Martins e F. Clemente)
Martins, F.
Statistical Significance Analysis of the R-DPSO (M. S. Couceiro, N. Ferreira, R. P. Rocha, F. M. L. Martins e F. Clemente), Proc ISEC - Symp. on Computational Intelligence for Engineering Systems, Coimbra, Portugal, Vol. 1, pp. CD-ROM - CD-ROM, November, 2011.
Digital Object Identifier:
Abstract
This paper presents a statistical significance analy-sis of a modified version of the Particle Swarm Optimization (PSO) on groups of simulated robots performing a distributed exploration task, denoted as R-DPSO (Robotic DPSO). This work aims to evaluate this novel exploration strategy studying the performance of the algorithm under communication con-straints while increasing the population of robots. Experimen-tal results show that there is no linear relationship between the number of robots and the maximum communication range. In general, the decreased performance by the developed algo-rithm under communication constraints can be overcome by slightly increasing the number of robots as the maximum communication range is decreased.