Research On Path Planning And Tracking Control Of Autonomous Vehicles On Slope Condition

With the escalation of people’s demand for a better life,the vehicle industry has gradually turned to intelligent development,and unmanned is one of the important directions of vehicle intelligence.The unmanned vehicle integrates the technology of planning and control,which has higher requirements for driving safety,system accuracy,stability,and drive experience.The road surface condition,complex and diverse,is one of the important influencing factors of the vehicle stability.When the vehicle is driving,it will inevitably pass through the slope and other areas.For unmanned vehicles driving on slope terrain,the performance of their planning and control directly affects the driving safety and stability.Therefore,it is necessary to study the path planning and control based on vehicle dynamics under the condition of slope to improve the autonomous driving ability of the vehicle.This thesis takes the four-wheel independent drive vehicle as the object,and mainly studies the path planning,dynamic model,tire driving torque distribution method and tracking control of the vehicle under slope conditions.Firstly,the path planning and optimization are studied under the condition of the slope.The efficiency and global convergence of the algorithm are improved by improving the artificial fish swarm algorithm.At the same time,adding the height factor as one of the optimization indexes reduces the frequent acceleration and deceleration of vehicles and improves economy and drive comfort.There are two steps for the smoothing of the planned path points which are found by the improved artificial fish swarm algorithm,including eliminating redundant points to shorten the driving distance and using the Bezier curve to connect path points to obtain a smooth and continuous curved path.Secondly,due to the uneven load distribution of the vehicle driving on the slope terrain,it is easy to cause the longitudinal and lateral sliding of the vehicle.In order to analyze the impact of slope road on the load distribution of unmanned vehicles,a three-degree-of-freedom vehicle dynamics model and an approximate linear tire model are established.To improve the real-time control,a tracking controller with real-time parameter updating is proposed.By updating the state coefficient of the controller in real-time,the controller can minimize the influence of external disturbance on the output of the result.The tire torque distribution is analyzed,and the torque optimization distribution controller which considering the slope is designed to improve the driving stability.By controlling the steering angle of the front wheels and the driving torque of each wheel,it can track the desired value of the lateral position,longitudinal position,yaw angle,and speed.The mathematical model simulation of the designed tracking controller verifies the correctness of the theory and the feasibility of the computer program,which tests two trajectories of the straight line and double shift line respectively.In this thesis,CarSim and Matlab are used to carry out union simulation experiments for slopes with different gradients.CarSim is used to establish a vehicle model with driving torques and wheel steering angles as external inputs,and the slope road environment.And a tracking controller based on linear variable parameter model predictive algorithm and torque optimization distribution algorithm is built in Matlab.The results show that the controller designed in this thesis can stably and effectively track the expected trajectory on the slope at different gradients.Compared with the controller without considering the slope factor,the larger the gradient of the slope,the better the control effect of the controller designed in this thesis.Finally,in order to verify the traceability of the path planned in this thesis,combined with the path planning part and the tracking controller part,the tracking simulation experiment is carried out on the planned curve with the proposed tracking controller.When the Bezier function takes the points according to the ordinary defined proportional coefficient,the length between the points taken on the curve will not be equal.For this problem,the proportional coefficient at the uniform time after correction is proposed to obtain the uniform motion points corresponding to the proportional coefficient.And ensure that the velocity of each point on the trajectory is relatively stable.The co-simulation results show that the path of the Bezier curve can be tracked by the proposed controller,and the tracking effect is good.