Research On Integrated Trajectory Planning And Control For The Intelligent Vehicle

Nowadays,when automobiles are highly widespread,there are frequent traffic accidents caused by human drivers themselves.The development of smart car technology is of great significance for reducing human-induced car accidents.Path planning and tracking control technologies determine the final driving behavior of a smart car,which are one of the key technologies of the modular autonomous driving system.This thesis proposes an integrated trajectory planning control framework for intelligent vehicles based on optimal control theory.Detailedly,the framework takes environmental information,vehicle state information and vehicle driving goals as inputs,and uses vehicle steering wheel angular rate and tire control torque as output control variables.This framework aims to solve the problems of discontinuous path and neglecting vehicle dynamics characteristics in planning layer,and low tracking accuracy of control layer that are common in existing hierarchical path planning and tracking control technologies.These problems ultimately lead to poor driving quality of the vehicle.The main research contents are as follows:(1)In order to balance the vehicle control accuracy and calculation cost under different working conditions,a variety of vehicle models and tire models have been built.1)In order to meet the vehicle dynamics characteristics under different working conditions,according to the principle of simple to complex,kinematic model,5-DOF and 7-DOF vehicle models are built in sequence.2)In order to accurately calculate the tire force,a linear tire model based on slip and a nonlinear tire model based on magic formulas are built,as well as a dynamic lagged tire model used to describe the dynamic characteristics of the tire.Carsim software was used to compare and verify the three vehicle models under the two test conditions of 1 Radian Step Steer and Sine with Dwell.The results show that the global trajectory errors of the three vehicle models are all within 0.2m.Among them,the 7-DOF model is the closest to the Carsim model.(2)In order to provide the initial vehicle state configuration for the integrated algorithm,a vehicle key state estimation algorithm based on EKF was designed.The EKF designed in this thesis uses the more common wheel angular velocity sensors to estimate the key vehicle states such as the vehicle’s lateral and longitudinal velocity in an iterative manner.The above-mentioned state estimation algorithm is verified on the Simulink software.(3)A vehicle integrated trajectory planning control algorithm is designed for the integrated framework.Design corresponding performance index functions according to different actual scenarios such as lane changing conditions,overtaking conditions and emergency obstacle avoidance conditions.Establish "vehicle-road" dynamic constraints according to the law of vehicle movement on the road.Establish driving restrictions based on road boundaries and obstacle avoidance conditions and other driving safety conditions.At last,according to the optimal control theory,the trajectory planning and control task of the vehicle is transformed into the optimal control problem.In this thesis,the direct shooting method is used to solve the above optimization problem offline.The results show that the proposed integrated trajectory planning control algorithm can obtain the optimal control variable sequence that conforms to the operating habits of human drivers.(4)A vehicle integrated trajectory planning control framework has been designed,and its online simulation experiments are carried out under three working conditions.The framework designed in this thesis mainly includes an integrated planning control module and a vehicle state estimation module.Among them,the vehicle state estimation module estimates the state information of the vehicle in real time according to the sensor input information,and the integrated planning control module optimizes the output variables based on the given vehicle state information.Based on the cosimulation platform,the feasibility of the integrated trajectory planning control framework designed in this thesis is verified.The trajectory plots show that the vehicle integrated trajectory planning control framework can effectively control the vehicle to complete the predetermined driving task under the conditions of lane changing,overtaking and emergency obstacle avoidance.