参数自优化的有人与无人车辆编队鲁棒模型预测控制

To solve the problem of disturbances in unmanned vehicle tracking control caused by the emergency acceleration, deceleration and steering control input of the manned leading vehicle in a formation of manned and unmanned vehicles, a parameter self-optimizing robust model predictive controller is designed. The noise extremum of the disturbances is determined by collecting and analyzing the historical data, which is scaled moderately to obtain a robust boundary. A local feedback robust controller is designed to restrain the disturbances, and the controller's parameters are automatically optimized using the Bayesian optimization algorithm. The mixed-integer linear optimization method is used to predict the trajectory of the leading vehicle, and a robust model predictive controller is proposed to track the leading vehicle using an unmanned vehicle. The simulation and experimental results show that the robust model predictive controller designed in this paper has a significant improvement in tracking accuracy compared with traditional controllers. The controller also effectively restrains the disturbances caused by emergency acceleration, deceleration and steering control input of the manned leading vehicle, model uncertainty of unmanned tracking vehicle and other external factors. Vibration is obviously suppressed, and the robustness of the system is enhanced.