多机器人编队系统体系结构的设计与性能分析  

Design of intelligent multirobot teams control architecture and its performance analysis

作　　者：王醒策[1] 顾国昌[2] 张汝波 宋梅萍[2] 

机构地区：[1]北京师范大学信息科学与技术学院,北京100875 [2]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2005年第3期311-315,共5页Journal of Harbin Engineering University

基　　金：中国科学院沈阳自动化所机器人学重点实验室基金资助项目(RL200106);国防科学技术工业委员会基础研究基金资助项目(51416070101CB0107).

摘　　要：根据多机器人智能编队的要求,设计分层多机器人体系结构.该结构主要有3层:协作任务层、协调行为层、动作控制层.协作任务层应用势场栅格法增强路径的安全性和可靠性并降低了计算复杂度;协调行为层应用强化学习法增强了机器人群体的智能性;动作控制层应用模糊控制实现了动作的匹配;通讯模块在各个层次上实现数据的传递.这种结构既能实现机器人个体的自主性和智能性,又能满足多机器人之间的协作与协调性的需要.每一层任务分配清晰,层层规划中都有智能体现,是一个真正的智能性结构.最后,仿真试验进一步验证了这种结构的有效性.此结构有很好的可扩展性,可以应用于相似的多机器人系统.A layered control architecture was designed for the formation of intelligent multirobot team. There were three levels: the cooperation task level, the coordination behavior level and the action planning level. The cooperation task level used a potential grid method to improve the safety of the path and reduce calculation complexity. The coordination behavior level used reinforcement learning to strengthen the robots' intelligence. The action planning level used fuzzy planning methods to realize the action matching. A communication model transfers messages between levels. This architecture not only shows the independence and the intelligence of a single robot but also the cooperation and the coordination among robots. In each level, the task was distributed reasonably and clearly. At last, the feasibility of the architecture was verified further through simulation. The expandibility of the architecture is good and the architecture can be used in similar systems.

关 键 词：多机器人 智能体系统 体系结构 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]