Agricultural Vehicle Based On CarSim/Simulink Path Tracking Control Simulation And Experiment

Path tracking control is an important part of autonomous driving technology.The application of path tracking technology in the process of automatic driving of agricultural vehicles can improve the operation accuracy and greatly improve the operation efficiency of intelligent agricultural vehicles,thereby replacing manual driving.At present,the path control of agricultural vehicles still has core problems such as low tracking accuracy and poor body stability.Therefore,this paper studies the lateral error and longitudinal speed control in the process of agricultural vehicle path tracking.The main contents are as follows:(1)Establish a mathematical model of a four-wheeled agricultural vehicle.After analyzing the operating environment and requirements of the vehicle,the kinematic model and dynamic model of the vehicle were established respectively.Based on the magic formula tire model,the changes of tire slip angle and slip rate under different vertical loads were verified in the CarSim simulation software.The simulation results show that when the slip rate is within ±5%,the tire longitudinal force and the slip rate have a linear relationship;when the sideslip angle is within ±10°,the tire lateral force and sideslip angle can be approximately regarded as a linear relationship,so in the case of small slip rate and slip angle,the tire is regarded as a rigid body,and its tire deformation under vertical load is not considered.(2)Design the lateral path tracking controller and longitudinal speed controller of agricultural vehicles.According to the kinematics model of the two-wheeled vehicle,a pure pursuit lateral path tracking control model is established,and the tracking effect of the pure pursuit control model and algorithm under low-speed driving of agricultural vehicles is verified by simulation;an error-based dynamic model of agricultural vehicles is established,and on this basis Design a discrete linear quadratic regulator path following control model.At the same time,in the design of the longitudinal speed controller,the upper-layer controller controls the longitudinal speed and acceleration of the vehicle based on Proportional Integral Derivative,and the lower-layer controller avoids frequent switching of vehicle driving states by formulating the switching logic of accelerator and brake.The construction of the lateral path tracking controller and the longitudinal velocity controller provides the basis for subsequent simulations and experiments.(3)Build a horizontal and vertical joint controller based on CarSim/Simulink.In the CarSim simulation environment,according to the actual vehicle parameters of the Dongfanghong LX954 tractor,the tractor model and the operation field simulation scene model are established,and the horizontal and vertical joint path tracking control model is established in the Simulink environment.Through the joint simulation of CarSim and Simulink,the tractor is realized.In the path tracking simulation under the straight path with slope and U-turn path,the simulation results show that the longitudinal speed controller can respond quickly and follow the desired speed as expected,and has a good control effect on the longitudinal speed;The control effect is better than the pure tracking controller in the straight line and turning process,but the robustness of the pure tracking controller is better than that of the LQR controller in the case of uneven roads in the field.(4)Build a real vehicle test platform to verify the actual control effect of the controller.The UWB positioning module is used to locate the test site,the inertial measurement unit is selected to obtain the vehicle speed and heading information,and the embedded development version is used to deploy the path tracking control algorithm program,to test the straight path and U-turn path.The test results show that: pure pursuit path control When the expected speed is 0.25m/s,the average tracking error of the straight section path is0.04 m,and the average error at the turning point is 0.21m;when the speed is 0.5m/s,the average tracking error of the straight section path is 0.05m;The average error of the linear quadratic regulator controller is 0.23 m at the speed of 0.25m/s;when the speed is 0.25m/s,the average error of the linear quadratic regulator controller is 0.026 m in tracking the straight-line path,and the average error at the turning point is 0.11m;when the speed is0.5m/s,the average error of tracking the straight-line path is 0.11m;The error is 0.05m;the average error at turns is 0.12 m.The tracking error of the straight line segment is not more than 4cm than the simulation error,and the tracking error of the turning point is not more than 7cm.It can be seen that the effect of the linear quadratic regulator controller is better than that of the pure pursuit controller in tracking the straight path and turning process,but the pure pursuit controller has better robustness,and the designed longitudinal velocity controller can track the desired path faster and at the same time stably to control the longitudinal speed.