Design of Path Following System Using Model Predictive Control with Adaptive Preview Time

• Main Authors: Cheng-Ting Tsai, 蔡承廷
• Other Authors: Bo-Chiuan Chen
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/mj5wv8

碩士 === 國立臺北科技大學 === 車輛工程系所 === 103 === Path following system can help the driver to maintain the vehicle driving along the target path for lane following control and automatic lane change control. The preview time of the model predictive control is often set to be a constant value. However, the preview time of the driver actually varies with the geometry change of road in front of the vehicle. Thus a model predictive control with adaptive preview time is proposed for path following control in this thesis. First, the steering actuator dynamics is included in the 5-state bicycle model for the controller design. A path geometry change (PGC) is proposed to adjust the preview time of the controller. The PGC is defined as the mean value of the average of absolute values of the first derivatives and the average of absolute values of the double derivatives of the target path in the prediction horizon. Model predictive control is used to generate the optimal steering angle command for tracking the target path. Simulation and experimental results show that the proposed algorithm can effectively reduce the path following error for tighter path tracking while reducing the lateral acceleration and jerk for better ride comfort.