Wavelet neural network-based intelligent transportation control system design

• Main Authors: TAI,HSIANG-CHU, 戴祥竹
• Other Authors: Ruey-Long Su
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/y78mf7

碩士 === 中國科技大學 === 資訊科技應用研究所碩士在職專班 === 102 === Car following on traffic safety has been an active area of research. However, human driving involves reaction times, delays, and human errors that affect safe driving adversely. One way to eliminate human errors and delays in car following is to replace the human diver with a computer control system. This thesis presents a wavelet neural network (WNN)-based control system for the car-following control. In this control system, a WNN is the main controller used to mimic a perfect sliding mode control law and a compensated controller is designed to compensate for the difference between the ideal perfect sliding mode control law and the WNN. The on-line adaptive laws of the control system, including the adaptive estimation law of approximation error bound, are derived based on the Lyapunov stability theorem and gradient descent method, so that the stability of the system can be guaranteed. Finally, simulation results show that this method can provide a safe car-following control.