机构地区: 北京理工大学自动控制系
出 处: 《计算机仿真》 2007年第1期148-150,221,共4页
摘 要: 在Robocup小型组(F-180)足球机器人比赛中,随着比赛对机器人运动能力和系统实时性要求的越来越高,基于本地传感器和处理器的主动避障策略日益受到人们的重视。模糊控制方法不依赖于精确的环境信息,并且运算量小,应用于自主避障很有前途,但传统的模糊控制方法难于手工创建控制规则,并且缺乏学习能力。研究了采用遗传算法对模糊路径规划器的模糊规则进行在线自动提取和优化的方法,并提出了一种用于评价规则有效性的适应度函数。仿真结果表明了算法有较好的适用性。 In the Small Soccer Robocup Game (F- 180), with the increasing requirement on the system real - time manoeuvre and the move capability, the self obstacle - avoidance, based on local sensor and local processor has gained greater importance today. Fuzzy control needs neither exact information of environments, nor huge amount of calculation, but it has the lack of study ability, and it' s not easy to set fuzzy rules of the system. In this paper, a fuzzy strategy for the self obstacle - avoidance of F- 180 soccer robot has been studied, and GA is used for fuzzy rules' on - line building, abstraction and optimization, and a fitness function is provided. The simulation result proves the applicability.
领 域: [自动化与计算机技术] [自动化与计算机技术]