Research On Lidar Guided Parking Technology With Combination Of Cyber And Physical

Safe parking is a classic problem of auto-driving cars.Autonomous Mobile Robots(AMR)have similar requirements for environment perception,path planning,and accurate parking when working in a local environment.Lidar is widely accepted in both car and AMR scenarios for guidance.The application of the virtual physical system method to the research of lidar-guided safe parking technology has significant advantages in the safety of the research and development process,the completeness of the code,and the richness of the scene.This thesis uses the method of virtual physical system to explore the problem of lidar-guided safe parking.The specific content is as follows:A research scheme of lidar guided parking technology combining virtual and real is proposed.On the basis of studying the common characteristics of common parking methods of vehicles and autonomous mobile robot parking scenarios,the parking application scenario model is refined;based on the research of the working principles of solid-state lidar and mechanical lidar,a safe parking guidance is proposed.Overall plan,and developed a research method combined with virtual physical system.Based on physical lidar,a research on safe parking technology is carried out.Based on the data of mechanical lidar,use clustering algorithm,feature point extraction and straight line fitting data processing methods to obtain key point coordinates,and complete safe path guidance by combining with the constraints of the kinematics model during vehicle parking;Based on the lidar data,the Lab VIEW software tool is used to realize real-time display of the parking environment;on the basis of processing the depth vision image,the obstacle contour display and collision avoidance warning function are realized.The virtual sensor technology of Coppelia Sim platform is studied.Based on the research of the mechanical characteristics,functional characteristics and communication methods of the two real lidars,the design and realization of virtual mechanical lidar and virtual solid-state lidar are completed.Finally,a simulation experiment platform was built to carry out simulation experiments.Complete the performance test of two virtual lidars;use python and Coppelia Sim co-simulation to carry out the vehicle safe parking guidance test.The experimental results show that the virtual lidar and the physical lidar have the same performance and successfully completed the parking test,verifying the virtual physical system Studying the feasibility of lidar-guided safe parking issues provides a novel development and verification method for intelligent applications such as autopilot and AMR.