Autonomous Navigation Technology Of Regional Intelligent Electric Vehicle Based On ROS

At present,artificial intelligence and other technologies are developing rapidly.In the new wave of science and technology,driverless technology has become a global industry innovation research hotspot,which has been paid attention by many national research institutions.Intelligent vehicles will become the inevitable trend of the future development of the automobile industry,not only that,driverless technology will also bring broad prospects for the future development of transportation.Autonomous navigation is the ultimate development goal of intelligent vehicles.This paper focuses on the research of real-time positioning,map building and autonomous navigation technology to realize vehicle intelligence.The main research contents are as follows:First of all,the process of building regional intelligent electric vehicle is analyzed in detail from two aspects of software and hardware.Taking the robot operating system ROS as the software platform and the intelligent electric vehicle as the hardware foundation,the whole experimental platform is built.The advantages and operation mechanism of robot operating system are summarized.Research the whole structure of intelligent electric vehicle,select multi-sensor,industrial computer and other hardware,complete the perception module,decision-making module and control module,and form a complete platform system.Then,the principle of lidar ranging is analyzed,and the key problems of simultaneous location and map building technology are studied.The slam technology based on extended Kalman filter and particle filter are studied respectively.In the process of traditional particle filter simultaneous location and map building,there is a problem that the increase of particle number makes the calculation quantity of the algorithm larger and the frequent resampling leads to the gradual dissipation of particles.The improved proposed distribution and the RBPF simultaneous location and map building method optimized by adaptive resampling technology are selected to realize the map building of unknown environment in simulation environment and real scene respectively.Finally,the framework and algorithm flow of autonomous navigation system are studied.The core of autonomous navigation is reasonable path planning,which is divided into global path planning and local path planning.It focuses on the global path planning algorithm based on A* and the local path planning algorithm based on TEB.According to the map information and the positioning and navigation framework,the relevant tests are completed on theregional intelligent electric vehicle to realize the autonomous navigation of the regional intelligent electric vehicle.