Fault Diagnosis Method for Mobile Robots Using Multi-CMAC Neural Networks
Multi-CMAC (Cerebellar Model Articulation Controller) neural networks based fault detection and diagnosis (FDD) method for mobile robots are proposed. Three failure types (system fault, sensor fault, and combined fault) are handled. Mobile robot system consists of several functional modules belonging to different module groups, which execute different tasks. According to the consistency among sensors information between the neighbor modules in the same module group, the method of fault diagnosis is studied. Then, multiple CMAC neural networks are used to implement the diagnosis. One CMAC neural network is set to one module group. In the neural network, the sensor information is used as the inputs and the fault signals are used as the outputs. As an example, the method is implemented on a drive system of a wheeled mobile robot. The simulation results show the effectiveness of the proposed technique.
fault detection and diagnosis fault types functional module mobile robot CMAC neural network
Yutian Liu Jingping Jiang
College of Electrical Engineering, Zhejiang University Zheda Road, Hangzhou, Zhejiang, 310027, China
国际会议
2007 IEEE International Conference on Automation and Lofistics
山东济南
英文
2007-08-18（万方平台首次上网日期，不代表论文的发表时间）