基于模型预测控制的无人驾驶履带车辆轨迹跟踪方法研究

Trajectory tracking control of unmanned tracked vehicle is challenged by complex system uncertainties and external disturbances. A kinematic model based on the instantaneous steering center is developed by studying the interaction between track and ground. Considering the fact that the reference path is a series of discrete waypoints, an adaptive reference path fitting method, which utilizes the third-order Bezier curve, is presented for path smoothing while providing road curvature information. Taking the unavoidable system uncertainty and external disturbance into account, a model predictive control based trajectory tracking controller with feedback correction is designed to systematically handle the modeling errors, environmental constraints, and actuator saturations. Real vehicle tests demonstrate that the proposed control scheme can be used effectively to restrain the effects of system uncertainties and external disturbances, while achieves the satisfying trajectory tracking performance of unmanned tracked vehicle.