Evaluation Of Signal Timing For Signalized Intersections Based On Multi-source Date Fusion

The traffic flow data detectiong is an important part of the signal control system.The intersection of urban road is the node of the urban road network.The effect of signal control not only determines the efficiency of the urban road network directly but also related to the evaluation of the urban traffic management and the smooth of traffic by various of travels such as drivers.However,there is a lack of quantitative evaluation of the effect of signal control at urban road intersections for a long time.This paper aims to obtain rich and efficienct traffic flow characteristics parameters by multi-source detection of traffic flow data fusion,and evaluate the operation quality of the signal control intersections quantitatively.The main content of this paper are multi-source traffic flow data fusion and the evaluation of the signal control intersection quality.According to the characteristics of traffic flow data,the overall framework based on the integration of data layer,feature layer and decision layer is designed to evaluate the traffic state and control performance of the road traffic signal control.The research mainly includes the following aspects:1?Aiming at the problem that the quality of multi-source traffic flow data,preprocess the data of fixed detectors and floating car and improve the quality of the traffic flow data so that to provide a good data foundation for data fusion.2?Based on the feature of multi-source traffic flow data and fusion framework,select appropriate data fusion algorithm.Established a traffic flow data fusion model of BP neural network and a velocity data fusion model of Bayesian estimation.Obtain traffic flow characteristic parameters according to multi-source detection data,laying the foundation for decision-level fusion.3?Evaluate the intersection operation.Select the appropriate index according to the uncertainty of the state of the signal control intersection and evaluate the state of the intersection by the weighted fuzzy comprensive evaluation.Propose a comprehensive evaluation index considering signal timing parameters and parking rate.And finally the feasibility of the algorithm is verified by example analysis.