Research On Traffic Flow Prediction And Control-Induced Synergy Based On Multi-Information Fusion

With the acceleration of urbanization process and the development of automobile industry,the contradiction between the capacity of existing urban roads and the increasing traffic demand has become sharp increasingly,and the phenomenon of traffic congestion has become prominent increasingly.It is a necessary part to realize intelligent traffic to analyze the influence and action mechanism of traffic flow characteristics on traffic congestion,construct a multi-information fusion model of traffic control and induced collaboration,and formulate effective strategies for traffic flow prediction and traffic control induced collaboration.It is one of the key technologies of intelligent transportation to master the traffic flow prediction data of the urban road network in real time.Fully grasping and analyzing the changing rules of the traffic flow can provide more accurate data protection for the coordinated research of traffic control and induction.This paper studies the progressive relationship of traffic flow analysis—traffic flow prediction—traffic control and guidance coordination,in which the spatial relationship between each intersection of the urban traffic road network,the history of each intersection,and the future traffic The law of flow data is described and studied in detail,so as to realize the prediction of traffic flow.Based on the prediction results,a traffic control and guidance coordination model is constructed to complete the traffic grooming.This paper conducts an in-depth study on these major issues,and validates the corresponding theoretical methods using actual on-site traffic data.The main research contents are as follows:(1)A traffic flow prediction model based on information fusion neural network(IFNN)is proposed.This model makes full use of the historical and future traffic information of road intersections to predict the traffic flow,which makes up for the shortage that the current model can only predict the output of the next moment based on the time sequence information of the past moment.(2)A road vectorization code is designed to vectorize road intersections,to reflect the spatial relation of road network with vector distance,and then the trajectory data containing spatial relation is input into the IFNN model for training and prediction.(3)Proposed a variety of traffic information fusion technology,comprehensively considering the historical data of the intersection,future data,real-time road traffic event data and fusion,combined with neural network algorithm for traffic data classification.On the basis of this,a traffic control and guidance collaboration model is completed in conjunction with the Brain Collaboration System(BCOS).Finally,the traffic simulation software VISSIM is used to real-time approximate simulation,and the superiority of this model is verified by the result evaluation.In conclusion,this paper has conducted a comprehensive and in-depth exploration on the problems of dynamic traffic flow prediction and traffic control guidance based on reconstructed track data,which has great reference and reference significance for solving the problem of urban traffic congestion.