| Unmanned ground vehicles can perceive the surrounding environment through vehicle sensors,generate the real-time paths using path planning algorithm,determine the best path by the decision-making system,and finally achieve path tracking under the action of vehicle controllers.In recent years,unmanned ground vehicles play an increasingly important role in the intelligent transportation system,on which many vehicle manufacturers and scientific research institutions have conducted extensive technical research and practical exploration.And trajectory tracking for unmanned ground vehicles has been studied as one of the research priorities.Aiming at the trajectory tracking problem of unmanned ground vehicles,the relevant control models and control strategies are put forward to realize vehicle trajectory tracking control.Firstly,in order to solve the problem of high-precision trajectory tracking for unmanned ground vehicles,an eight-degree-of-freedom vehicle dynamic model and a two-degree-of-freedom vehicle dynamic model are established,which are used to accurately characterize the dynamic changes of the states of the vehicle.Those models of vehicle lay the foundation for the design of the model-based trajectory tracking controllers.At the same time,CarSim simulation software is used to verify the accuracy of the vehicle dynamics models.Secondly,based on the Model Predictive Control(MPC)algorithm,trajectory tracking controllers are proposed to solve the multi-state variable constraint problem in vehicle trajectory tracking control.The controllers facing the vehicle posture error model realizes the posture error tracking of vehicle.At the same time,compared with the traditional Lyapunov function which is selected more difficultly,MPC algorithm can be easier to achieve the controller design.Thirdly,the hierarchical coordination control strategy is proposed for the trajectory tracking problem of unmanned ground vehicles,which involves a multi-level control of kinematics and dynamics.The trajectory planning layer utilizes the MPC to realize the dynamic trajectory planning.The intermediate control layer uses Sliding Mode Control(SMC)to keep the dynamic trajectory real-time tracking,and the torque distribution layer reasonably allocates the wheel torques.The proposed hierarchical coordination control strategy is verified by the simulation of vehicle autonomous lane-changing control.Finally,in order to solve the problem of uncertainties in the process of vehicle trajectory tracking,a controller based on Radial Basis Function(RBF)neural network is proposed to realize the dynamic estimation and compensation of vehicle uncertainties.The effectiveness of the RBF neural network on the uncertainty estimation is verified by the simulation of the control system.The robustness of the controller with RBF neural network compensation is verified by comparison simulation of the trajectory tracking with uncertainties.The MATLAB / Simulink simulation platform is used to model the vehicle trajectory tracking control system,and to verify the effectiveness of the above control methods and control strategies. |
