An alliance generation algorithm based on modified particle swarm optimization for multiple emotional robots pursuit-evader problem

This paper researches the alliance generation algorithm with emotional factors on the basis of multiple robots pursuit-evader problem. Firstly, this paper constructs an emotional model for pursuit robots: we not only apply the basic emotion method to the emotional expression, but also simulate the process of emotional transfer with Hidden Markov Model (HMM). Secondly, we determine the cooperation intention according to the robots´ emotional factors, so that we can prevent the robots with the negative emotions from involving in the mission in case of a negative impact on the alliance. Then, we introduce the subgroup size on the foundation of particle swarm optimization (PSO) to avoid the premature convergence problem, thus the algorithm can obtain the maximum profit in a relatively short period of time. Finally, we bring in the dynamic redistribution mechanism for a better pursuit efficiency.