City expressway traffic state rolling prediction method based on fusion data

Abstract

The invention relates to a city expressway traffic state rolling prediction method based on multi-source heterogeneous fusion data; the method comprises the following steps: aiming at a to-be-predicted road section and obtaining block port and microwave data of detection points; obtaining true travel time of matching vehicles passing the block port, using a microwave detector to obtain the real speeds of the detection points, and carrying out multi-source heterogeneous data fusion according to space relative positions; using a non-linear interpolation algorithm to obtain the speed of a random non-detection point, and building a virtual trajectory to obtain an estimated travel time; optimizing the target function (root-mean-square error of speed and travel time) so as to calibrate parameters; using calibrated parameters to predict the speed and travel time of the next time period. The method is simple and effective, easy to operate, thus preventing the problems that a conventional method cannot be changed in a parameter calibration process once parameters are calibrated.

Description

technical field [0001] The invention relates to the field of traffic state prediction, in particular to a traffic state rolling prediction method based on fusion of multi-source heterogeneous data. Background technique [0002] In urban road traffic, the traffic flow is participated by tens of thousands of people and vehicles. Each traffic participant will make decisions in line with his own interests based on his own behavior goals and codes of conduct, and according to different traffic operating environments. . This kind of decision-making mainly comes from the estimation of the operation status of the traffic flow in the future. Only by accurately grasping the behavior changes during the operation of the traffic flow can the decision-making made be more targeted. Therefore, taking the dynamic prediction of traffic flow as an important content of daily monitoring and management is the basis of traffic management decision-making science and the key to active and all-round...

AI Technical Summary

Problems solved by technology

The method of the invention is simple and effective, easy to operate, overcomes the deficiency that the traditional static parameter calibration method cannot be applied to real-time traffic state prediction, and solves the limitation of the application range of the existing traffic parameter calibration method

Images