| The data processing and analysis of driving simulation experiment are helpful for the experimenters in the field of traffic engineering to conduct in-depth research on driver behavior.However,the existing driving simulation experiments have huge data and complicated experimental procedures,and the subjects vary from person to person,but the basic data processing methods and analysis methods are roughly the same.In order to facilitate subsequent analysis,scholars in the field of transportation often need to learn different programming languages to segment and extract the original simulation data,which undoubtedly adds repeated research steps.In addition,a large amount of data based on time points is not intuitive,so it is time-consuming and labor-consuming to find effective key areas manually.Therefore,starting from software engineering,this paper combines driving simulation experiment and data comprehensive analysis method,taking experimenters and driving simulators as research objects,systematically analyzes the problems that need to be solved in the data and comprehensive management of driving simulation laboratory,and develops a data processing and analysis platform for driving simulation laboratory.It is designed to facilitate the experimenter in the field of transportation to extract special analysis indicators,visually display driving simulation data,comprehensively evaluate drivers and comprehensively manage the driving simulation laboratory.This paper analyzes the current situation of driving simulation and behavior evaluation at home and abroad.On the basis of the existing driver behavior evaluation,the subjective questionnaire analysis method and radar chart comprehensive evaluation method are proposed to analyze and demonstrate the evaluation.Then,on the basis of explaining the system theory and related technology,the detailed functions and implementation methods of the system are determined.In terms of system function design,the data processing and analysis platform of the driving simulation laboratory is divided into four modules: data storage,data processing and display,driving data analysis and laboratory comprehensive management.In terms of architecture design,SSM background framework and Vue.js front-end framework were selected for separate development of front and rear ends,My SQL relational database was used,Redis was used as cache,and Tomcat server was used for system development.In the aspect of system implementation,the driving simulation data are classified and stored,and the ECharts tool provides the function of custom data extraction and visual display,which can visualize the univariate trend and multivariate relationship.In terms of driver comprehensive evaluation,factor analysis and K-means clustering were used to process the collected questionnaire information,and the driving behavior classification based on subjective questionnaire was obtained.The radar chart was used to make a comprehensive and intuitive quantitative evaluation by using the comprehensive scoring method of "control-decision-avoidance".In the integrated management module,the experimental process is sorted out to standardize the management.Finally,based on the design,the system is tested,summarized and prospected. |