Research On Model Predictive Control Of Intelligent Vehicle Trajectory Planning And Tracking

Intelligent vehicles are the results of human social science technology,and it is an important frontier field that combines traditional automobile industry.The intelligent vehicles can seriously improve the safety of motor vehicle driving as a necessary part of the intelligent transportation system.The purpose of energy saving and environmental protection is gained.Trajectory planning and tracking control is the important part of that intelligent vehicles drive safely.The thesis concentrates on the trajectory planning and tracking of intelligent vehicles.The controller is designed to realize intelligent vehicle trajectory planning and tracking control with using the Laguerre function and adding the planning layer based on model predictive control.It is aimed to improve system performance.The main research contents are as follows:1.The lateral controller combined with the Laguerre function is designed based on the vehicle kinematics model.At First,the Laguerre function is introduced.Then the intelligent vehicle kinematics model is linearized,and a lateral controller combining model predictive control and the Laguerre function is designed to optimize the objective function.And it is iterated to a constrained standard quadratic programming problem,and solved.At last,the superiority and stability of the lateral controller are verified by MATLAB.2.The trajectory tracking controller combined with the Laguerre function is designed Based on the vehicle kinematics model.At First,the vehicle kinematics model adds a external disturbance.And the control method combining feedforward and feedback controller combined with the Laguerre function is used.Then the model with added disturbance is linearized and discretized.The objective function with control increment constraint is optimized by introducing the Laguerre function.It is as the constrained quadratic programming problem.It is solved with iteration.The Lyapunov theorem is used to analyze its stability.At last,the superiority and stability of the trajectory tracking controller are verified by MATLAB.3.The obstacle avoidance vehicle trajectory tracking control system with added planning layer is designed based on the vehicle dynamics model.At First,the state-space equation is built and converted.The objective function is formulated using the Laguerre function,and vehicle dynamics constraints are added to the system to improve the stability of tracking.Then the real-time optimal scrolling problem with constraints is solved and the optimal solution is found.The planning layer is added to the trajectory tracking layer to form a two-layer control system combining trajectory re-planning and tracking control.In the planning layer,the nonlinear model prediction algorithm and the obstacle avoidance function are combined to achieve reliable obstacle avoidance.The method of selecting trajectory points on the reference trajectory is improved,and a method of fusing obstacle information is used.The curve fitting method based on the fifth-order polynomial is used to fit the planned trajectory points,which can simplify the transfer parameters between the control layer and the planning layer.It has a high accuracy.At last,the verification and analysis are performed by Car Sim/Simulink.