基于实际误差函数和隶属函数机器人避障算法  

Algorithm about mobile robot evading obstacles in path planning based on an error function and a membership function with actual physical content

作　　者：张寒松[1] 贾瑞清[2] 王廷军[2] 

机构地区：[1]中冶京诚工程技术有限公司炼钢工程技术所,北京100176 [2]中国矿业大学机电与信息工程学院,北京100083

出　　处：《辽宁工程技术大学学报（自然科学版）》2006年第4期588-591,共4页Journal of Liaoning Technical University (Natural Science)

基　　金：国家自然科学基金资助项目(50475094)

摘　　要：介绍一种机器人在不确定环境下,通过探障传感器,在探测过程中采用一种模糊神经网络来躲避障碍物的方法。该模糊控制采用有明确意义的模糊隶属函数,由采用有明确物理意义误差评价函数的神经网络来实现,以避免模糊控制的死点从而完成避障路径规划。计算机仿真实验的结果验证了该方法的有效性和实用性。This paper introduces a kind of evading obstacles method of mobile robot which is in an uncertain environment, and the mobile robot can evade the obstacles detected by means of a kind of fuzzy ANN algorithm. And the fuzzy control adopts the membership function which has the real physical content, and it is realized by ANN which adopts the error function with the real physical content, so the blocking zone of fuzzy control can be avoided and the obstacles can be evaded. Then the emulation experiments show that this method is efective and practicable.

关 键 词：机器人 避障 模糊控制 隶属函数 神经网络 误差函数 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP242[自动化与计算机技术—控制科学与工程]