摘要
移动机器人系统由多个功能模块组成,每个功能模块实现不同的功能。本文根据各模块之间传感器信息的连贯性,定义了一种故障分类方法,将移动机器人的故障分为系统故障、传感器故障和混合故障三类。依据该分类方法提出了一种基于多个CMAC神经网络的故障诊断处理方法。该故障诊断方法的基本思想是将处理后的各模块组的传感器信息作为神经网络的输入,故障类型作为输出,利用CMAC神经网络完成各模块组的故障诊断过程。最后,仿真实验的结果证明了该故障诊断方法在移动机器人故障诊断上的可行性。
Mobile robot system consists of several functional modules. According to the consistency of the sensor information between the neighbour modules, a method of classifying the faults was put forword. All the faults are classified into system fault, sensor fault and combined fault. A muhi-CMAC neural network based fault diagnosis method was described. In this method, the sensor information work as the inputs and the fault types as the outputs, then the CMAC neural network are used to finish the process of fault diagnosis. The simulation results show the effectiveness of the proposed technique.
出处
《机床与液压》
北大核心
2007年第11期169-173,共5页
Machine Tool & Hydraulics
基金
国防科技预研基金资助(00J16.6.3.JW0401)
关键词
故障分类
故障诊断
功能模块
移动机器人
CMAC神经网络
Fault classification
Fault detection and diagnosis
Functional module
Mobile robot
CMAC neural network
