Research On Relative Abnormal Driving Behavior Of Vehicles Based On Floating Car GPS Data Analysis

With the continuous improvement of people's living standards and the rapid development of the automobile industry,the number of drivers and car ownership in our country are continuously increasing.At the same time,road traffic safety is also facing increasing pressure.The number of traffic accidents that occur each year in China is as high as millions of times,causing heavy losses to people's lives and property.Although the causes of traffic accidents are varied,human factors still occupy a dominant position.Drivers gradually develop fixed driving habits during the period of accumulating driving,and the driver's driving behaviors reflected by these habits can have a potential impact on road traffic safety.Numerous case studies also indicate that the driver's abnormal driving behavior is the main cause of many traffic accidents.Therefore,it is of great significance to study the abnormal driving behaviors of drivers in order to improve road safety and reduce the loss of people's lives and property.Floating car data was collected from various vehicles equipped with on-board GPS positioning devices,and the use of floating car data to obtain road traffic information is one of the main technical means commonly used by research scholars of many countries in recent years.The methodological conclusions derived from the research of floating car data can be applied to private cars,key surveillance vehicles,and dangerous vehicles.Limited by current technical and objective factors such as storage and network,floating car data used for driving behavior analysis is often sparse.For sparse floating car data,this paper uses scientific calculation methods to reduce the error caused by data sparsity to experimental results and correct the abnormal data.At present,there is a lack of uniform classification criteria and quantitative indicators for abnormal vehicle driving behavior in China.Most studies only classify driving behavior by simply setting a fixed threshold.These method lacks in-depth analysis of floating car data,and the data source and quality requirements are too harsh.After analyzing and correcting the data of real floating cars,this paper first proposed the concept of relative abnormal driving behavior,establishes the relatively abnormal driving behavior model.Then,based on this,this paper proposes reasonable relative abnormal driving behavior identification and evaluation methods to distinguish the relative abnormal driving behavior between vehicles.Through experimental analysis,the weighting coefficient of the three relative abnormal driving behaviors in the total score was determined.At the same time,through the correlation test,the validity of the model and method was proved.The research methods proposed in this paper can be used in insurance,advertising,public welfare and other fields in the future,such as adjusting the rate of vehicle insurance expenses according to the difference in the relative abnormal driving behavior scores of different vehicles;or establish a reward mechanism for drivers with better driving behaviors to encourage drivers to maintain good driving habits.In particular,the analysis of the behavior of dangerous vehicles and taxi drivers can greatly reduce the major traffic accidents caused by abnormal behaviors of drivers.