Trajectory Planning And Tracking Control For 4WS Autonomous Vehicle

Lane-changing is the most common scenario in the transportation system,and its execution effect has a significant impact on the efficiency and reliability of the transportation system and the safety and comfort of passengers.With the continuous deepening and popularization of autonomous vehicle technology,lane-changing process of intelligent vehicles has increasingly become a research focus.The development of Vehicle-to-Vehicle(V2V)collaboration technology provides more advanced strategies to the research of lane-changing,especially in the process of trajectory planning,and also accelerates the realization of smart transportation and smart cities.At the same time,more intelligent vehicles also need more flexible mechanical structures while the research objects in related fields are mostly traditional front-wheel-steering vehicles at present,which lack the forward-looking integration with more complex four-wheel steering(4WS)vehicles.Therefore,this paper uses the four-wheel-steering intelligent chassis as the carrier to study the lane-changing process of four-wheel-steering autonomous vehicle,establishing a safe,efficient and comfortable lane-changing trajectory planning and tracking control program.The main work of this subject is as follows:1.Four-wheel-steering intelligent chassis modelingThe system architecture and functional assembly of the four-wheel-steering intelligent chassis are introduced and the dynamics model of the vehicle is analyzed in detail.Four-wheel-steering two-degree-of-freedom vehicle dynamic model suitable for the chassis is established on this basis and the rear wheel angle control is added to accurately reflect the motion response in the case of simultaneous input from the front and rear wheels and input from the front wheel alone.2.Research on trajectory planning algorithm for V2V sceneBased on the study of the behavior of vehicle lane-changing process,the characteristics and advantages of the V2 V system are analyzed combined with the framework of the V2 V system.An ideal lane-changing scene is constructed,and the fifth-order-polynomial curve is used as the trajectory planning method to obtain surrounding traffic information to determine the constraints of the starting and ending positions of lane-changing,and the feasible trajectory interval can be obtained.3.Obstacle avoidance and optimization algorithm researchThe extended rectangular vehicle model is established to implement the obstacle avoidance processing for the feasible trajectory interval.Taking the lane-changing efficiency and passenger comfort as the influencing factors of the optimization function,the optimal planned trajectory can be obtained.The system updates the vehicle information in real time to dynamically adjust the planned trajectory to ensure the overall safety,efficiency and comfort of the planning trajectory.4.Research on trajectory tracking control based on MPCThe allocation strategy of front and rear wheel angles of four-wheel-steering vehicles is analyzed,and a four-wheel-steering trajectory tracking controller based on Model Predictive Control(MPC)algorithm is designed.According to the planned reference trajectory,reference heading angle and other reference inputs,the wheel angle of the four-wheel-steering intelligent chassis is calculated in real time under the premise of ensuring safety and stability to achieve tracking control of the target trajectory.5.Algorithm integration and simulation testBy designing the test conditions and programming in the PyCharm environment,the trajectory planning,obstacle avoidance and optimization algorithm simulation verification is implemented according to the relevant parameters of the four-wheelsteering intelligent chassis and road information.The driving route and status of each vehicle are simulated and dynamically displayed.The target trajectory parameters are derived,the vehicle dynamics model is built through MATLAB/Simulink platform,and the four-wheel-steering intelligent chassis trajectory tracking controller is simulated to verify the feasibility of the algorithm.The research significance of this paper is that it proposes an autonomous lanechanging process trajectory planning,obstacle avoidance and optimization program,which is more convenient and efficient than traditional research by reducing the cost of technology landing and promoting intelligent transportation.The Forward-looking fusion research with the structure of four-wheel-steering vehicles is conducted and the four-wheel-steering trajectory tracking controller is constructed through the model predictive control method.The one-stop simulation test design from trajectory planning to tracking control promotes the full-stack development of autonomous vehicle technology.