Research On Prediction Model Of Operating Speed For Passenger Cars On Expressway By Considering The Drivers’ Visual Area Of The Road Surface

In recent years, China has made breakthrough progresses in the field of expressway construction, but at the same time the traffic safety problem has become more and more severe. Among the factors that threaten traffic safety and induce traffic accident, the rationality of road alignment design has a great influence on the traffic safety. The road alignment on the one hand, to ensure that the vehicle can normally travel, on the other hand, have to pay attention to the continuity of the linear, balanced and coordinated, in order to ensure traffic safety. Safety evaluation method of domestic and foreign current is commonly similar to the Guidelines for Safety Audit of Highway and the most core content of the safety evaluation system is the determination of the vehicle’s operating speed.80% information of the expressway passed to the driver through the visual way, after the analysis of driver’s central nervous system to make the corresponding decision to control the speed behavior, in order to maintain the vehicle for safe driving. Therefore, with the help of UC-WINROAD simulation software and the basic theory of driver’s vision, the author puts forward the comprehensive of driver’s visual road surface area represent road’s flat, vertical and horizontal parameters, and according to driver’s visual road surface area to study the operating speed of passenger car.Taking the pixel as the basic unit of measuring driver’s visual road surface area and using the driver’s visual road surface area as evaluation index, the influence of road plane, longitudinal section and transverse section on the running speed of passenger cars is considered synthetically. First in the experimental stage a multi-functional data acquisition instrument which integrates the technology of video collection, the technology of GPS and the technology of high precision speed pick up was used for gathering the drivers’ visual area of the road surface and operating speed. Then the collected single frame color vision image was gray scale processed using linear transformation method and got a gray field image, through the discontinuity and similarity of the pixel gray value, the improved Otsu algorithm is composed to carry out the binarization segmentation of road table and the background, and got the binary image. At last, based on the pixel line scan algorithm, scanned binary image from the bottom to extract the driver’s visual road surface area. SPSS software was used to analyze the correlation between visual road surface area and speed of drivers through scatterplots of qualitative analysis, Pearson, Kendall and Spearman quantitative analysis, and established a mathematical relationship between the two.Finally, through the analysis of examples and the comparison between the predicted values and the measured values of the model, the accuracy, reliability and validity of the prediction model were verified.