Research On Trajectory Tracking Control For Autonomous Intelligent Platform With Four-wheel Steering

As a key technology in the intelligent vehicle,trajectory tracking control has been extensively studied.The currently control methods are mostly applied to front-wheel steering(FWS)vehicles,rather than four-wheel steering(4WS).Therefore,this paper proposed the advanced trajectory tracking control strategies for intelligent mobile platforms,considering the characteristics of 4WS architecture.This research is funded by National Natural Science Foundation of China(51705185)-Intelligent collaborative control of All-terrain vehicle via active attitude and 4WS control systems,where the topics mainly focus on the investigations of FWS and 4WS trajectory tracking control strategies.The simulation results of different strategies are analyzed under specific reference trajectories,including lane changing maneuver,constant curvature maneuver and sigmoid loops maneuver.Simulation results show that the two different controllers designed for the platform can both track the reference trajectories.Besides,the 4WS trajectory tracking controller has higher tracking accuracy than the FWS trajectory tracking controller under the same reference trajectories.Moreover,its adaptation to the longitudinal speed of the platform is more prominent as well as the performance in the extreme conditions.The experimental results of the mobile platform also present that both controllers can track the reference trajectories and the performance of the FWS trajectory tracking controller is better under specific conditions.Furthermore,the experimental data are close to the simulation results,verifying the consistency of the simulation and experiment with the favorable effectiveness of the control strategies.The main research contents of this paper are organized as follows:1.Modeling of 4WS platformBased on the detailed analysis of the classic vehicle dynamics model and the characteristics of the driverless intelligent mobile platform,4WS dynamics model is established,which is suitable for the trajectory tracking control of the platform by introducing the additional control input of the rear wheel angle.Here,some parameters in the FWS model are modified to ensure that the proposed model can accurately reflect the motion response of the plant when the simultaneous inputs are carried out for both front and rear wheels.2.Research on FWS trajectory tracking controller based on MPCA mature model predictive control method is used to calculate the front wheel angle of the platform in real time yielding to the safety and stability,where the controller operates by using the reference inputs such as reference trajectory and heading angle,so as to track the preset trajectory.The designed controller takes small calculation burden.The model can be directly applied to real vehicle experiments after compilation with little dependence on hardware conditions,where the good coordination is achieved between the calculation burden and the trajectory tracking accuracy.3.Research on 4WS trajectory tracking controller based on adaptive-MPCIn order to make better use of the 4WS advantage of the platform,an adaptive-MPC algorithm is proposed here to develop a 4WS trajectory tracking controller.The adaptive MPC algorithm overcomes the limitations of the linear MPC on the parameters of the dynamic model.The adaptive-MPC algorithm can update the structure of the platform dynamic model in real time while the transition matrix changes with the longitudinal speed variation of the platform.During the calculation process,the latest platform dynamics model is internalized into the prediction model of MPC.Therefore,it shows better adaptability to longitudinal vehicle speed variation.4.Experimental verification of trajectory tracking strategiesClosed-loop experiments of trajectory tracking were carried out on the ground based on ultra-wideband wireless(UWB)communication.First,the fundamental performances of the intelligent mobile platform were tested,including acceleration performance,braking behavior and steering accuracy.Subsequently,the hardware arrangement of the trajectory tracking control system was integrated,where the OXTS RT2502 inertial navigation system and the MicroAutoBox1513/1514 controller were installed.In terms of software,MATLAB/Simulink and ControlDesk were used to verify the different control strategies mentioned above and applied to the platform after compilation.The experimental results demonstrate that the designed controllers can achieve the expected performance on the platform.Meanwhile,the dynamic model of the platform is also well verified.The influence of the rear-wheel steering for the lateral stability is mainly investigated in this research,the designs of different trajectory tracking controllers are proved suitable for the platform especially for the intelligent movement of 4WS systems.It is worth mentioning that the trajectory tracking experiments of the platform further reveal the excellent effectiveness of the control strategy and the consistency with the simulation results.