Research On Trajectory Tracking Controller Of Unmanned Vehicle Based On Planar Four Wheels

In the driver-vehicle-road driving system,the driver is often faced with many uncertain factors.Due to factors such as fatigue,lack of driving experience,and misjudgment of road conditions,safety traffic accidents occur frequently.With the continuous advancement of cloud technology,5G communication,and automatic control technology,the autonomous driving technology has become a hot research topic in the automotive field.There is an urgent need to solve the problem of stable and accurate trajectory tracking in the control system.This article takes a planar four-wheel unmanned vehicle as the research object and uses vehicle kinematics models,dynamics model,and tire model as the basis.Based on model predictive control theory and introducing lateral angle constraint conditions,a trajectory planner and tracker based on linear time-varying model predictive control are designed to control the vehicle’s active front-wheel steering and track the reference trajectory.This article focuses on the study of planar four-wheel unmanned vehicles and establishes a vehicle kinematics,dynamics,and tire model that considers longitudinal,lateral,and yaw degrees of freedom.The "Magic Formula" formula is used to derive the tire model,and the dynamic characteristics of the tire under pure slip and pure side slip conditions are obtained.The tire is analyzed through simulation to validate the changes in longitudinal and lateral forces.To address the complexity and real-time requirements of controlling nonlinear systems,the nonlinear system is transformed into a linear time-varying system.The Linear Time-Varying Model Predictive Control(LTV-MPC)algorithm is designed and derived.Transforming the Optimization problems into QP problems to prevent the occurrence of infeasible solutions,a relaxation factor is added to the objective function.Considering the safety and stability requirements of high-speed driving,a soft constraint on the lateral deviation angle is added to the LTV-MPC algorithm to prevent the vehicle from losing its ability to track the trajectory due to front-wheel slip while turning on a curved road.The study utilized Simulink and Car Sim to create a joint simulation platform.A suitable vehicle model and tire model were selected in Car Sim,and the LTV-MPC trajectory tracking control algorithm was encapsulated in S-function form to simulate and verify the performance of the designed controller.Based on the vehicle kinematic model,three initial speeds of 3m/s,5m/s,and 10m/s were set,and two reference trajectories of straight line and circular were selected for simulation testing of the trajectory tracking controller’s performance at low speeds.Based on the vehicle dynamic model,the performance of three different controllers was compared under low road surface adhesion coefficient conditions,with reference to the double moving line at three different initial speeds of 10km/h,20km/h,and 30km/h.The experimental results showed that the trajectory tracking controller designed in this study can adapt to changes in vehicle speed and has high accuracy and stability.Considering the complexity of the vehicle driving environment,obstacle avoidance trajectory planning was added on the basis of trajectory tracking.A hierarchical control system was established,with an upper-level trajectory planner with obstacle avoidance function and a lower-level trajectory tracking controller.The performance of the hierarchical control system was verified under low road surface adhesion coefficient conditions with a single obstacle and double obstacles in the reference trajectory and at different initial speeds.The experimental results showed that the vehicle could successfully avoid obstacles under different speed and obstacle conditions,and the real-time performance of the hierarchical controller was good.