Research On Evaluation Algorithm Of Driving Behavior Of Commercial Vehicle Drivers

With the rapid development of social economy and logistics industry,road transportation is still the most extensive mode of transportation.China’s commercial vehicle market,especially the commercial truck market is increasingly prosperous.The driver is the most important part of the "human-vehicle-road" transportation system,and the quality of the driver’s driving technology is greatly related to the economy and safety of the vehicle.Therefore,in this context,it is increasingly important to study the driving behavior of commercial vehicle drivers.In this paper,a vehicle driving condition recognition method based on hierarchical decomposition method and fuzzy recognition method is proposed,and a quantitative evaluation algorithm is proposed for different recognized driving conditions,and a scoring model of driver’s driving behavior is constructed.The specific work of the paper is as follows:(1)Preprocess the collected real vehicle data,reduce the noise in the collected data and eliminate the abnormal data.The hierarchical decomposition method is used to decompose the vehicle condition into driver’s driving intention recognition and vehicle driving state recognition.Then the fuzzy logic toolbox in MATLAB is used to design the fuzzy controller of driver’s driving intention recognition and the fuzzy controller of vehicle driving state recognition respectively,and the simulation model of vehicle condition recognition is built in MATLAB/Simulink.Finally,the model is validated with real vehicle data.(2)Set the evaluation parameters of each working condition of the vehicle,and propose corresponding evaluation algorithms and quantitative scoring standards for different working conditions identified,such as start,acceleration,braking and deceleration,uphill and downhill.For example,for starting and accelerating conditions,the difference between engine torque change and engine ideal torque change is scored.For uphill and downhill conditions,the change of acceleration is evaluated by comparing.The braking deceleration condition is evaluated by its deceleration size and continuous braking time.In addition,the paper also studied the engine economic speed utilization rate,shift duration and the matching relationship between engine speed,speed and gear,and other driving level evaluation algorithm.(3)A scoring visual program is designed by using App Designer in MATLAB and verified by real car data.The program can output the driver’s driving level evaluation score in different driving processes,which can evaluate whether the driver’s driving behavior is good or not.The results of this study can roughly judge the bad operation of the driver in the driving process and urge the driver to correct by analyzing the driving data,so as to achieve the purpose of saving fuel consumption and reducing the incidence of accidents.