Building Method For Indoor Intelligent Vehicle Test Conditions

Traditional Hardware-in-the-Loop(HIL)testing often uses simulation software to build virtual test conditions,which ignores many practical factors,such as there is no real vehicle in the test conditions,and the existence of the radar reflection surface is ignored,which is not easy to ensure the accuracy of the test results,although outdoor field test can reflect the real traffic environment,the investment cost is huge.An indoor intelligent vehicle test platform can not only solve practical problems,such as no real traffic participating vehicles under a virtual test condition,but also has a relatively low investment cost.In this laboratory,it is necessary to have a platform that can quickly realize the building of multiple test conditions.Therefore,this article studies the building method of indoor intelligent vehicle test conditions,which can realize the rapid construction and application of test conditions.First,aiming at the problem of vehicle path tracking control,the basic principles of Pure＿Pursuit,Stanley and Model Predictive Control were introduced.Based on algorithm principle,three path tracking controllers were designed.The Simulink and Car Sim cosimulation method was used to verify the correctness and tracking performance of the controller.The double-line shifting condition was set for the tracking experiment,then the applicable speed ranges of the three path tracking controllers were given.By adding multiple vehicle lateral dynamic models and controlling them through the path tracking controller,the simulation platform of the test conditions was built in the Simulink environment.Using Raspberry Pi as the development board,a PID speed controller was designed to control the longitudinal speed of the smart car.The speed information was collected through the encoder and the heading angle of the smart car was collected by the LSM9DS1 IMU sensor,then calculated position information.According to the position information,the path tracking controller calculates the turning angle of the front wheel at the next moment,then realized the path tracking function of the smart car and the control code was automatically generated and deployed.In the same local area network,the upper computer can start the code located in the lower computer,then realized the start and stop control of multiple smart cars.The smart car tracking experiment results show the correctness of the path-following controller and the feasibility of control method for multiple cars.Finally,on the basis of the above work,a simulation and control software for indoor intelligent vehicle test cases was designed by using MATLAB/GUIDE.The software can not only simulate the test conditions,but also realized the development of the path tracking control code to the lower computer.At the same time,it can control the vehicle and monitor vehicle status.