Dynamic Path Tracking Control Under Vehicle And Road Constraints

As a global automotive engineering research hot spot and a new growth engine for the automotive industry,intelligent vehicles are a comprehensive system that integrates environmental perception,planning and management,decision,as well as assisted driving on several levels.Intelligent vehicle motion control is the key point to achieve autonomous driving.In order to make the vehicle in accordance with the expected planning of the trajectory of safe driving,driving decisions have to be made based on the current surrounding environment and the movement of the vehicle body,vehicle speed and other information,and motion commands have to be sent to acceleration pedal,brake pedal and steering actuator.As a typical complex dynamic coupling system,the motion control of intelligent vehicles is not only related to the vehicle itself,but also to factors such as roads and driving scenes.How to design an efficient motion control method in a complex transportation system consisting of vehicles,roads and environments is the goal and the difficulty in the field of intelligent vehicle research.In response to this complex problem,a method of horizontal and vertical coordinated motion control is presented to adapt to vehicle and road constraints,and to solve the problems such as tracking deviation and tail slaying of a single control under complex working conditions.Simulation experiments and hardware-in-the-loop experiments are carried out to verify the effectiveness and real-time performance of the tracking algorithm under various working conditions.The main research contents include the following parts:(1)Establishing the vehicle coordinate system and geodetic coordinate system realizes the conversion of the vehicle position coordinates,while considering the non-linear characteristics of the vehicle,the lateral and longitudinal vehicle dynamics models are established respectively,and the dynamics of the vehicle is analyzed.Based on the simulation of the nonlinear tire model,the relationship between the tire grip and the road grip coefficient is analyzed,which provides a model basis for the design of the lane following control system.(2)The vehicle lateral tracking controller is designed based on the theory of optimal control.The Frenet coordinate system is established to calculate the tracking error between the vehicle position and the desired path.The two degrees state space equation is established,and the cost function is built to solve the problem.The Riccati equation is solved to obtain the optimal LQR feedback control gain.At the same time,the feed forward angle compensation control is designed to eliminate the steady state error of the system.The efficiency and reliability of this control method are verified by simulation.(3)For transverse tracking at high speed and low adhesion,it is easy to appear tracking deviation or tail slashing.The expected longitudinal speed of the vehicle is calculated based on the fuzzy control theory and the limit of adhesion force by considering the factors such as road adhesion coefficient,road curvature and vehicle lateral tracking error,so as to ensure the optimal adhesion of the wheel.Based on the theory of predictive model control,the upper vehicle longitudinal velocity tracking controller is designed,and the desired acceleration is achieved by constraining the acceleration and acceleration increment.Further,the underlying controller is designed based on the reverse engine model,and the throttle and brake pressure are controlled according to the engine torque required to achieve the vehicle speed following control.Considering yaw phenomena such as oversteer and understeer while following the vehicle,a lane-following stability controller is established and additional braking torque is applied to generate an additional yaw moment to eliminate lugging yaw angle error,to ensure that the vehicle works on extreme roads.It is safe to drive in conditions.The efficiency of the designed controller is verified by CarSim/Simulink co-simulation.(4)In order to further verify the applicability and real-time of the designed coordinated control algorithm,a hardware-in-the-loop simulation experiment platform is built and the secondary development has been achieved.Real-time communication is achieved among CarSim / Simulink and the NI-PXI real-time machine and controller.The Simulink model is compiled and downloaded to the NI-PXI real-time simulator,and the simulation scene is established based on CarSim,and the simulation experiment is carried out in the simulation road.Experiments show that the lateral tracking and longitudinal velocity planning and tracking designed in this academic dissertation are both effective and real-time,and can run effectively in real hardware.