改进的粒子群优化算法在机器人足球中的应用  

Improved algorithm of particle swarm optimizationand its application in Robot Soccer

作　　者：张彦铎[1] 田晖[1] 

机构地区：[1]武汉化工学院计算机科学与工程学院,湖北武汉430073

出　　处：《哈尔滨工业大学学报》2005年第7期905-908,共4页Journal of Harbin Institute of Technology

基　　金：湖北省教育厅科学研究重大项目资助(2002020003);湖北省优秀中青年创新团队项目资助.

摘　　要：为了提高算法对复杂动态环境的适应性,提出一种改进的粒子群优化算法,在确保其收敛性的前提下,采用动态跟踪优化的方法,获得动态环境中相对较高的性能.并将该方法应用于机器人足球比赛中,解决了多个机器人间的竞争与合作,以及机器人最佳动作的选择问题,实验结果表明该方法是有效的.As a representative method of swarm intelligence optimization, Particle Swarm Optimization is an algorithm for searching the complex space efficiently through cooperation and competition among the individuals in a population of particles. For the sake of enhancing the applicability of particle swarm optimization algorithm to dynamic environment, a modified algorithm is proposed. The method ensures the astringency, employs the strategy of tracking and optimizing the dynamic system to achieve relatively high performance. The algorithm is applied to the problem of robots' cooperation and competition and the selection of robots' best action in robot soccer. Experimental results show that the method is effective.

关 键 词：群体智能 粒子群优化 机器人足球 多智能体 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]