Research On Traffic State Discrimination Method Based On Traffic Radar Data

Traffic congestion has been affecting the country’s economic development and people’s normal life for a long time,effective identification of traffic status is the premise of solving congestion,and the collection of traffic information provides an important judgement basis for traffic state discrimination.Aiming at the problems that the traditional detection equipment of intersection is easy to be affected by weather and the accuracy is not high,this paper proposes to use the traffic radar to collect the road information to identify the traffic status,and choose the intersection as the research object to study the traffic state discrimination method based on the characteristics of traffic radar data.The main work of this paper is as follows:(1)Designing the application program to obtain the data of traffic radar in real time,using socket network programming technology to receive the radar original message,resolving the data according to the specified protocol to get the useful information of the targets,and analyzing the characteristics of the traffic radar data.At the end,the paper compares the resolved data with the GPS data to verify the correctness of the data acquisition application program.(2)According to the characteristics of radar data,analyzing the problems of traffic radar data,identifying the wrong data to preprocess the data,selecting queue length,traffic flow and average road speed to build the traffic condition discrimination index system.Designing a queuing length estimation model,a traffic flow estimation model and a road average speed estimation model based on the traffic radar data characteristics.The validity of the estimation model of the discrimination index is verified by the measured data of the intersection.(3)Through analysis,the fuzzy C-means(FCM)algorithm is used to identify the traffic state of the intersection.Using the cluster validity CH index to determine the number of clusters;In view of the problem that FCM algorithm is easily affected by the initial clustering center and falls into local optimum,so genetic algorithm is used to optimize the initial clustering center;According to the different contribution of each discrimination index to traffic state,the final weight of the discrimination index can be obtained by combining expert scoring and analytic hierarchy process(AHP)in subjective weighting method and Relief F method and entropy weight method in objective weight method,the traditional FCM algorithm is improved by euclidean distance weighted.Compared with the two algorithms by the measured data of the intersection.,experimental results show that the improved algorithm is more suitable for traffic state recognition.(4)Through the recognition of the intersection state,it is verified that the results of the improved algorithm in this paper are consistent with the actual traffic state of the intersection.Compared with the changes of traffic state on weekdays and weekends,the algorithm can identify different traffic states quickly and effectively;compared with the traditional single indicator recognition,the result of this algorithm is more accurate and stable.Finally,Win Form technology is used to design a visual interface for radar data,which can show the changes of road traffic status and traffic parameters.In this paper,the traffic radar data is used to establish a traffic state discrimination model,which is verified to be able to accurately and quickly identify the changes of the intersection state,it can lay a foundation for the realization of advanced traffic state management at intersections,such as advanced signal control systems,congestion guidance systems and so on to ease traffic congestion.