作者: Micael S Couceiro , Rui P Rocha , Nuno MF Ferreira , Patricia A Vargas , None
关键词: Swarm robotics 、 Artificial intelligence 、 Evolutionary computation 、 Evolutionary algorithm 、 Computer science 、 Communication complexity 、 Robot 、 Overhead (computing) 、 Population
摘要: The Robotic Darwinian Particle Swarm Optimization (RDPSO) recently introduced in the literature has ability to dynamically partition whole population of robots based on simple “punish-reward” rules. Although this evolutionary algorithm enables reduction amount required information exchange among robots, a further analysis communication complexity RDPSO needs be carried out so as evaluate its scalability. This paper analyses architecture system, thus describing dynamics data packet structure shared between teammates. Moreover, set rules is also proposed order reduce overhead within swarms robots. Experimental results with teams 15 real show that methodology reduces overhead, improving scalability and applicability algorithm.