Hybrid Control based on a Novel Fast Convergence and High Precision CMAC for Electric Loading System

摘要：

A novel CMAC hybrid control strategy with fast convergence and high precision is proposed for the electric loading system of UA Vs in order to solve disturbance of the surplus torque. This CMAC control strategy employs a nonuniform memory quantization scheme that defines its computing structure. A weighted Gaussian neighborhood-based activation process is subsequently implemented to facilitate the learning and computation of this novel CMAC structure. The hybrid controller is compound of CMAC and PD algorithm. The feedforward control is realized by CMAC while the feedback control is performed by PD controller. The mathematical model of an electric loading system for UAVs is established and the detailed control structure is put forward. Simulation results show that the proposed controller has a faster convergence rate and higher precision compared with the conventional CMAC, and can effectively eliminate the surplus torque and fairly improve the dynamic loading performance of the system.

关键词： electric loading system surplus torque CMAC neural network hybrid control quantization point Gaussian weighting.

作者: Bo Yang Zhe Wang

作者单位: School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China

会议类型: 国际会议

会议名称: 2009 9th International Conference on Electronic Measurement & Instruments(第九届电子测量与仪器国际会议 ICEMI2009)

会议地点: 北京

会议语种:英文

页码: 2318-2323

在线出版日期: 2009-08-16（万方平台首次上网日期，不代表论文的发表时间）