Driving Behavior Analysis Based On Beidou Vehicle Positioning Data

With the continuous and rapid economic growth and the improvement of the quality of life of our country,the car ownership is increasing year by year,and the road freight industry shows a vigorous development posture.However,the incidence of traffic accidents is also increasing,especially for the road operation vehicles that need long-distance driving,such as two passengers and one danger,safe travel is more important.Driver's driving behavior is the most important and direct cause of traffic accidents.In order to standardize the driving behavior of drivers,effectively monitor the driving status of vehicles,improve the safety of travel and facilitate the management of vehicle enterprises,the driving behavior analysis topic based on Beidou vehicle positioning data is proposed.Based on the positioning data of Beidou vehicles and relying on the dynamic monitoring platform of Beidou vehicles,a driving behavior analysis method based on random forest and K-means is developed in this paper.The vehicle driving monitoring system and vehicle driving behavior analysis system are designed and implemented,and the driving behavior analysis report of drivers is formed through the comprehensive analysis results of the system.The main research contents are as follows:(1)Design of driving monitoring system based on Beidou vehicle positioning data.Through analyzing the original location data,the driving state of the vehicle is analyzed.The driving state is visualized in the form of data and charts,and then analyzed driver's driving behavior.Consist of speed change,the use of lights and brakes,driving area analysis,drift data analysis,etc.,enterprises can real-time understand the driving information of vehicles.(2)Driving behavior analysis method based on k-means.K-means clustering algorithm is used to analyze the clustering of speed,latitude,longitude,light,braking and other data in the original data,as well as the clustering of various features in different time periods,and the clustering results are used as the basis for driver driving behavior scoring.(3)Driving behavior analysis method and Implementation Based on random forest.The original data is statistically processed and labeled,and the driving behavior analysis model is constructed by using random forest algorithm.The driving behavior of drivers can be classified according to the real-time data.Combining the analysis method with the monitoring platform,the driver is classified and scored according to the driving state data and K-means algorithm,and the driving behavior analysis report is formed.The relevant management department of the enterprise where the vehicle is located can master the driving status and driving behavior information of the vehicle in real time through the vehicle driving monitoring system and driving behavior analysis system,and carry out rewards and punishments or warning education and other corresponding measures according to the driving type and comprehensive score of the driver in the driving behavior analysis report,so as to achieve the purpose of safe driving.