Predictive Control And Software Implementation Of Unmanned Vehicle Path Tracking

In recent years,with the rapid development of the economic level,driving a car has become one of the modes of travel.The increasingly complex traffic environment has further stimulated the demand for safety,energy saving,environmental protection and comfort.At the same time,the development of technologies such as network,sensing and intelligent control has also promoted the development of driverless technology.At present,unmanned vehicle has been concerned by researchers in many fields.The path tracking control based on vehicle motion control is the basis of unmanned driving,so it has great research significance.In view of the current research on vehicle path tracking control,the mainstream strategy is to decouple the control of the vehicle’s on lateral and longitudinal directions.The longitudinal control is the vehicle speed control.The lateral control is the steering angle control.Current control issues are mainly focused on lateral control.This paper focuses on the implementation of lateral path tracking control.To design the controller by using the model predictive control method,and to implement the software of the path tracking control of the unmanned vehicle path tracking.The details are as follows:Firstly,the kinematics model is established in the ENU.Then the dynamic characteristics of the vehicle are analyzed to establish a linea r lateral dynamic model.The dynamic relationship between the steering angle input of the front wheel and the vehicle pose change can be described using the kinematics model and dynamics model.Based on the high-precision vehicle simulation software veDYNA,the simulation results show that the established model can meet the needs of controller design in the prediction horizon.The predicting model is used to predict the future motion characteristics of the vehicle,and the predictive control method is used to design the path tracking controller.Aiming at the slow reducing speed of lateral error and the steering angle oscillation,a constrained predictive control method for reducing steering angle oscillation by constrained control sequence increment is given.Aiming at the too long time of the model prediction control online optimization in practical application,the explicit model predictive control is studied,and the online calculation speed is improved by offline optimization calculation and online table lookup method.The simulation results show that the proposed method is feasible and effective.Aiming at the software implementation problem of the path tracking control of the unmanned vehicle,the software implementation of the path tracking control of the driverless car is carried out by constructing the localization and control module.The localization module can measure the position and the attitude of the vehicle;the control module can calculate the optimal control sequence according to the lateral error between the vehicle and the desired path based on the designed control algorithm.