Design And Implementation Of Automobile Anti-collision Warning System Based On Jetson Nano And Machine Vision

With the increase of the number of cars,traffic safety is becoming an increasingly serious issue.Against such a background,the anti-collision warning system is getting more and more attention from related enterprises and researchers.The very purpose of the system is to reduce the occurrence of traffic accidents and ensure vehicles’ safety.The paper has taken intelligent anti-collision warning system as its researching subject,on the basis of current active safety technology,the intelligent system can take preventive measures prior to the occurrence of accident so as to prevent such accidents from happening as much as possible.Based on sensor technology and image processing as well as Jetson nano embedded technology,this paper designs and implements a anti-collision warning system to assist the driver.The system consists of three parts,namely,main control data processing center,multisensor information collector and output information execution unit.Over the entire working process,at the outset,multi-sensor collects information concerning surrounding environment,followed that the main control processing center analyzes the solicited data,and determines the driving safety level as per the pre-set algorithm,finally the system takes appropriate measures as per the information and analysis.During the process of driving,the system can can predict such possible risks as lane deviation,front and rear vehicles,and obstacle collisions etc.,and give the driver warning and automatic manner,so as to avoid traffic accidents.Firstly,after analyzing the functional requirements of forward collision warning system,this research has designed an overall implementation scheme,dividing driving status into different safety levels;by using the image processing method to detect the lanes in the road image,the system can distinguish the practicable area;meanwhile,the deep learning method is used to detect the vehicle.Through experiments on the self-built data set and the KITTI data set,acute detection results can be achieved.Moreover,in this system,the lane deviation warning model is determined,besides,a longitudinal safety warning model suitable for the system is also established.Secondly,this research has designed hardware for the system,including the hardware selection and design of each sub-module;On the basis of hardware,design of software for each part;Finally,the system has been tested,including single module test and overall function test,the test results indicate that the overall design of the system is feasible,and thus,the target requirements of the research subject is satisfied.