Threshold Adaptive Adjustment And Optimization Of Traffic Incident Detection Algorithm

At present,road traffic is facing serious problems,such as traffic congestion and traffic incident.As one of the basic technologies of ITS,traffic incident detection play an important role in easing traffic congestion,improving traffic safety and ensuring travel efficiency.Due to the strong randomness of traffic incident and the complexity and variability of road traffic environment,the detection of traffic events is difficult.However,the traditional incident detection algorithm based on fixed threshold comparison is difficult to adapt to the real-time changes of road environment,so it is particularly important to develop a traffic event detection algorithm that can adaptively adjust the threshold.Based on this,combined with the actual accident data,traffic flow data,meteorological data and road data,this thesis established the threshold update conditions of the incident detection algorithm,and further determined the realtime update strategy of the threshold.On this basis,an adaptive traffic incident detection algorithm based on threshold adjustment and optimization was proposed.The main research content of this article includes the following aspects:Firstly,in order to decrease repeat alarms,the original traffic incident detection algorithm is modified by constraining the alarm time.Secondly,threshold adaptive adjustment and optimization method is established.By analyzing the impact factors on the performance of the traffic incident detection algorithm,threshold update conditions were established.Taking the performance index of the incident detection algorithm as the objective function,the threshold optimization model of traffic incident detection is established.Then,the particle swarm optimization algorithm is used to solve the model,obtain the optimal threshold parameters,and complete the determination of the initial threshold of the traffic incident detection algorithm.On this basis,combined with real-time data,the threshold optimization time window is introduced to complete the real-time adjustment and update of the threshold.Thus,an adaptive SND incident detection algorithm,an adaptive DES incident detection algorithm and an adaptive incident detection algorithm based on Kalman filter prediction are proposed.This thesis proposed an adaptive threshold adjustment and optimization method for the classic incident detection algorithm,forming an adaptive traffic incident detection algorithm,which is of great significance for improving the adaptability of the incident detection algorithm.