Research On Driving Behavior Safety Evaluation Method Based On Vehicle Trajectory Data Of "Two Kinds Of Buses And A Kind Of Vehicle Carrying Dangerous Goods"

With the rapid development of transportation,the number of "two kinds of buses and a kind of vehicle carrying dangerous goods" operating vehicles has significantly increased,posing significant challenges to road traffic safety.Therefore,based on the trajectory data of "two kinds of buses and a kind of vehicle carrying dangerous goods" vehicles,this thesis uses a multi-attribute evaluation method to evaluate the safety of risky driving behaviors on the road,providing a theoretical basis for road management departments to carry out traffic control and improve the transportation efficiency of vehicles.The main research contents are as follows:(1)Firstly,taking Hebei Province as the research area,this thesis elaborates on the source and concept of "two kinds of buses and a kind of vehicle carrying dangerous goods" trajectory data.In response to the problems of data redundancy,missing fields,and incorrect data in vehicle trajectory data,data processing is carried out to supplement the missing driving distance and speed in the trajectory data,facilitating further exploration of hidden driving behavior characteristics in the data.(2)Secondly,based on a large amount of trajectory data,the risky driving behaviors recognition algorithm is established using Python to quantify eight risky driving behaviors: over speed,rapid deceleration,rapid acceleration,vehicle jitter,rapid lane change,bad idle,stalling and taxiing and fatigue driving.On the one hand,conduct characteristic analysis on the frequency changes of daily and hourly risky driving behaviors;On the other hand,research is conducted on the speed changes of four risky driving behaviors: over speed,rapid deceleration,rapid acceleration,vehicle jitter,are studied to lay a foundation for determining the subordination function below.(3)Then,the multi-attribute TOPSIS evaluation method is used to evaluate the safety of risky driving behaviors.Firstly,the membership function combining ascending semi trapezoidal-Sigmoid is constructed to determine the evaluation indicators of risky driving behaviors.For each evaluation indicator,fuzzy set membership function analysis is performed,and the subjective weighting method and entropy objective weighting method are combined to subjectively and objectively fuse the evaluation indicators,Combining the advantages of TOPSIS and grey correlation evaluation method to achieve safety evaluation of risky driving behaviors.(4)Finally,the safety evaluation was conducted on the "two kinds of buses and a kind of vehicle carrying dangerous goods" vehicles in Hebei Province within a week,and the rationality and effectiveness of the multi-attribute TOPSIS safety evaluation method were tested using examples.Propose countermeasures and suggestions for the safe driving of "two kinds of buses and a kind of vehicle carrying dangerous goods" vehicles from three aspects: government,enterprises,and road facilities.By processing and identifying the trajectory data of "two kinds of buses and a kind of vehicle carrying dangerous goods" vehicles,a complete safety evaluation method for risky driving behaviors are constructed to achieve the classification of risk levels for one week’s driving behavior.The evaluation results are in good agreement with reality,and the effectiveness and feasibility of the combination weighting method and multi-attribute TOPSIS evaluation method in driving safety evaluation are verified.The research results can provide reference and guidance for government regulatory authorities and relevant enterprises to regulate the safe driving behavior of "two kinds of buses and a kind of vehicle carrying dangerous goods" operating vehicles.