Design Of Control-By-Wire System For Autonomous Driving And Its Application In Research

The intelligent mobile platform integrates computer,automatic control,sensors and vehicle engineering.It is expected to replace manual driving and reduce traffic accidents.The chassis-by-wire system is composed of steer-by-wire,drive-by-wire,brake-by-wire and it is the basis for intelligent platform to achieve vehicle horizontal and vertical control.The research on the design and application of the platform chassis wire control system is of great significance.In this paper,a completely set of wire control system is built for an electric vehicle chassis,and its application is studied.The communication system and power supply system equipped with the platform is discussed.Taking the electric chassis as the platform and completed the selection calculation and modification design of the bottom drive mechanism.The hardware of the whole platform controller is developed with Freescale xep100 as the main chip.The hardware system includes switch data acquisition,analog data acquisition,can bus communication,RS485 communication,PWM out,Wireless remote control command collection and other modules.The bottom code and application layer function library of the vehicle controller are compiled.The automatic control of the bottom layer’s execution modules is realized.Based on the Simulink platform,the kinematics simulation model of the autonomous platform is built.Taking the velocity and the average steering angle speed of the wheel as the control variables.steering kinematics characteristic is built and take the platform speed and the average steering angle speed as the control variables to study the steering range.The maximum curvature of the platform turning and the turning range at different velocity and steering angle speeds are calculated to study the turning control range of the platform,which lays a theoretical foundation for the study of steer-by-wire.Taking reverse thinking.Based on the kinematics simulation model,the initial area of the single-step vertical parking is researched.The various constraints of the single-step parking are analyzed.including initial position condition,parking space size condition,parking constraints condition,obstacle constraints condition,platform parameter constraints condition,etc.Based on the B-spline curve,the parking path optimization function is established,and finally design a parking path that satisfied the multiple nonlinear constraints of the parking process.Based on the lidar sensor to establish a ranging coordinate system,the weight coefficient of the obstacle distance and the weight coefficient of the speed function in the dynamic window is revised according to the distribution of the obstacles in the front of the vehicle.The results show that the improved dynamic window algorithm can effectively avoid obstacles and reach the target position in a short time.In order to verify the effectiveness of the established simulation model.In the square of the campus.Based on the modified wire control experimental platform,the maximum curvature and control range of the platform turning are actually tested and compared with the simulation model.The results show that the error between the actual platform test and the simulation calculation is about 0.13m.and the built kinematics model is effective.Based on the vertical parking space behind the main campus building to establish a parking space coordinate system.The target path data points calculated by the simulation is converted into GPS coordinate points and input them to the path tracking controller to complete the single-step vertical parking test.The results show that the system has high reliability,the parking track error is within allowable range,and the designed parking path based on the B-spline is effective.