| At present,China’s highways have been built on a large scale with the ETC(Electronic Toll Collection)system,which is one of the largest Io T systems in the world.The ETC system records all the driving trip tracks of highway ETC vehicle users,as well as related information.The highway ETC system is an advanced transportation Io T system that integrates computer processing technology,communication technology,control technology and information technology to achieve efficient,safe and reliable traffic management and provide more convenient services for traffic travel.Daily information exchange between vehicles driving on the highway and the ETC gantry system generates hundreds of millions of ETC transaction data.Data mining and analysis of the massive ETC transaction data is conducive to improving the safety management and congestion management of highways.In view of the problem that abnormal driving behavior of vehicles with ultra-low speed is likely to cause traffic congestion and traffic accidents,ETC transaction data is used to study the driving behavior of vehicles,establish the evaluation model of vehicles with abnormal driving behavior based on fuzzy comprehensive evaluation method,label ETC vehicle users with vehicles with abnormal driving behavior,and complete the construction of user portrait.The highway supervisory department can combine with ETC system to achieve the application goal,supervise the vehicles with abnormal driving behavior,monitor their abnormal driving behavior at ultra-low speed,and punish them when the situation is serious,which has certain significance and value for highway traffic safety and congestion management.The specific research work is shown below:(1)The abnormal detection of the ETC gantry system will generate the corresponding abnormal ETC transaction data.For the three types of repeated detection,missed detection and error detection,we propose the corresponding identification and processing methods,carry out abnormal transaction data cleaning,and make statistical analysis of the abnormal detection of the gantry in order to carry out the next research work.(2)Combining ETC transaction data and ETC gantry topology data table,explain the principles and ideas of constructing the riskiness evaluation index system for driving behavior,constructing evaluation indexes such as average speed,speed stability,average speed over speed times on road sections,average speed under speed times on road sections,and the specific construction process of evaluation indexes.(3)The evaluation model of vehicles with abnormal driving behavior is constructed based on the fuzzy comprehensive evaluation method.The evaluation is based on the riskiness evaluation index system of driving behavior of vehicle users,and the subjective weights of evaluation indexes are calculated using hierarchical analysis method,the objective weights of evaluation indexes are calculated using CRITIC assignment method,and the comprehensive weights are calculated based on game theory.The affiliation function of the evaluation indexes is constructed using box-line diagrams.(4)The practical application of the driving behavior abnormal vehicle evaluation model,get the driving behavior abnormal vehicle set,for which the ultra-low speed abnormal driving behavior situation is analyzed,the ultra-low speed abnormal driving behavior vehicle for example analysis verification.According to the pass speed smooth road section,as well as the ultra-low speed abnormal driving behavior vehicles in the smooth road section,combined with ETC video data,it is found that the ultra-low speed abnormal driving behavior vehicles do drive slowly in the leftmost overtaking lane,without keeping a safe distance from the vehicle in front of them,with a major suspicion of causing traffic congestion and traffic accidents,verifying that the driving behavior abnormal vehicle evaluation model constructed in this paper has certain validity and feasibility,and can indeed detect abnormal driving behavior at ultra-low speed. |
