摘要
针对Agent个体学习与群体学习各自的不足,探讨了多智能体系统(MAS)中的合作与协商及Agent学习技术,提出了一种新的混合智能学习算法.将个体学习与群体学习有效结合起来,提高了Agent的个体性能及系统整体的智能水平.在足球机器人仿真系统中进行了实验,结果表明了算法的可行性与有效性.
Multi-agent system (MAS) is a mayor branch of distributed artificial intelligence and one of its research focuses is agent learning algorithm. To solve more complex problems appeared nowadays, cooperation and negotiation together with Agent learning technique in MAS were studied, and a hybrid intelligent learning algorithm was proposed to combine individual learning and group learning and improve agent individual ability and the system intelligence level. The specific application and experimental results in robot soccer prove the feasibility and validity of this algorithm.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2003年第9期1083-1085,共3页
Journal of Harbin Institute of Technology
基金
山东省自然科学基金资助项目(Y2002G18).
