Urban Traffic Road Risk Assessment And Accident Risk Prediction Under Edge Computing Environment

With the rapid development of economy,the total number of vehicles in the country and the number of urban roads are also increasing,and more road traffic problems(traffic congestion,traffic accidents,etc.)also follow.The occurrence of urban traffic accidents will directly affect the traffic conditions of the road section where the accidents occur,leading to traffic congestion and even causing more serious losses.How to effectively predict road risks through traffic data,so as to alleviate traffic congestion and reduce traffic accidents has become a hot issue to be solved in the development of modern cities.Previous researchers to the prediction of traffic accidents are mostly a city historical data of traffic accident,to use a different model or improvement of traffic accident forecast model for a certain area,however,this kind of method of prediction and long-term only rule of traffic accident is relatively consistent,not in real time for road accidents have a better forecast.The occurrence of traffic accidents has a great relationship with the real-time traffic situation on the road.With the continuous development of intelligent devices,the group crowd sensing technology of the Internet of Vehicles and edge computing have become a hot topic to solve this problem.They can realize the analysis of vehicle data on the road surface with low delay and high efficiency.Different from the previous research on traffic accident prediction,this paper analyzes and forecasts the risk of urban traffic accidents on the basis of edge computing environment,combining real-time road traffic data information and historical traffic accident data information.The specific innovations are described as follows:1.A new road risk assessment method based on edge computing is proposed.The road vehicle information is collected by the group crowd sensing equipment,and the low delay and high reliability of the edge computing technology are used to monitor the road vehicle data and analyze the abnormal vehicle information and abnormal traffic flow information.Based on this,a risk assessment model is established to evaluate the road traffic risk,and then the minimum risk Bayesian decision method is used to verify and judge.In order to verify the effectiveness and practicability of this method,this paper conducts an experimental assessment of road risk on artificial road network and real road network respectively,and establishes a comparative experiment of route selection.Experimental results show that the proposed method is more secure than the traditional path selection method.2.An urban traffic accident risk prediction method based on edge computing is proposed.On the basis of the road risk assessment model based on edge computing,the traffic risk in the future of urban area is predicted by combining with the historical traffic accident data,which makes the prediction result more reliable under the influence of real-time traffic flow factors.First group of intellectual perception technology is used for road traffic information collection,the margin calculation was carried out on the road to real-time risk assessment,based on historical traffic accident data and combining with both Long and short term memory networks predictive model to an area of city traffic accident frequency to forecast,finally through a variety of prediction model of comparative experiments verify the validity and superiority of this method.The experimental results show that the prediction results of the model proposed in this paper are more accurate than other prediction models.