Research On Traffic Signal Coordination Control Method Based On Data Fusion

In recent years,China's economy has maintained steady growth,and the number of motor vehicles has continued to increase,resulting in increasingly serious traffic congestion.The vehicle-road cooperation system and automatic driving technology have become an important way to solve traffic congestion and improve traffic efficiency.However,the traffic signal control method based on the fixed detector data can't meet the signal control requirements of the future era of vehicle networking and vehicle-road coordination.In recent years,the mobile data acquisition technology such as GPS floating car is becoming more and more mature and widely used.There is a certain degree of redundancy and complementarity between the detection data of different detectors.In this paper,a data fusion method is proposed to process the data to obtain more accurate and comprehensive data.The traffic signal control parameters and control system are optimized and improved to provide data and decision support for traffic signal control.Firstly,this paper analyzes the development process and research status of data fusion in the field of transportation.By comparing the characteristics of different types of traffic data detectors and the requirements of different traffic control modes for detection data,the necessity of data fusion is explained,and the goal of data fusion is proposed.Then,the original data is preprocessed,such as removing and replacing.The data are fused by weighted average and BP neural network.Fitting the trajectory data and recalculating the delay time of the intersection,the Webster equation is improved.Considering the actual traffic conditions,the optimization method of green signal ratio in the case of congestion and non-congestion is given.In view of the MULTIBAND model's insufficient consideration of the red light queuing dissipating time,this paper fully consider the influence of the red light queuing dissipating time.The operation parameters of scoot control system,such as traffic capacity STOC,travel time JNYT,and the maximum queue clearing time QCMQ,are calculated to replace the manual investigation or estimation data and improve the control accuracy of the system.Finally,the hardware in the loop simulation system is built,and the SCOOT parameter optimization method is proposed in this paper.Taking the intersection along Zhongshan Road in Dalian as an example,the operation parameters of the SCOOT system are optimized and simulated.Two evaluation indicators of the number of stops at the intersection and the delay time at the intersection are selected.Compared with the original control scheme,the effectiveness of the scheme is proved.