A method for predicting route travel time based on RFID data

Abstract

The invention discloses an RFID (Radio Frequency Identification Device) data-based route travel time predicting method. The method comprises the steps of by taking a moment, when a guidance vehicle arrives at the beginning section of a guidance route, as a starting point of the predicted travel time, mining a historical RFID database, dividing the guidance route into a plurality of 'target road segments', respectively predicting the travel time of the road segments through time mode recognition models, and finally adding the travel time of the road segments to obtain the travel time prediction value of the guidance route. According to the method, RFID data can be effectively mined, the route travel time of a single vehicle can be effectively predicted, and the RFID data-based route travel time predicting method is different from the traditional average travel time-based route travel time predicting method. The method and technology adopted in the invention are simple and easy, operation conditions can be easily met, and the method is easy to popularize and apply in large and medium-size cities.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, in particular to the prediction of route travel time in the field of traffic flow guidance and management, and more specifically, to a method for predicting route travel time based on RFID data. Background technique [0002] Accurate route travel time prediction value is an important decision-making parameter in the traffic guidance system, which helps the system to find the optimal route (the shortest time route), and helps the driver to adjust the driving route so that the traffic flow distribution of the road network reaches the desired state , to achieve a balanced distribution of road network traffic flow, can improve traffic congestion to a large extent. [0003] Traditional route travel time refers to the arithmetic mean of all effective vehicle travel times passing through the induced route within the prediction interval. However, the service object of route guidance i...

AI Technical Summary

Problems solved by technology

[0009] (2) The detection data communication is susceptible to electromagnetic interference, and the positioning accuracy of the detection vehicle is not high (15-20 meters);

[0010] (3) In practice, the data sampling interval of the probe car is relatively large (about 20-30s);

[0011] (4) The detected data belongs to "point data", which cannot describe the traffic state of the road network in a wide range, accurately and comprehensively;

[0012] (5) The probe vehicle has a large spatio-temporal distribution inhomogeneity;

[0013] (6) When processing floating car data points, due to certain errors in the electronic map, the accuracy of travel time estimation and prediction based on floating car data is not high

[0019] (4) The number of vehicles equipped with RFID electronic tags is relatively large, and the sample size is sufficient;

[0021] However, even in large and medium-sized cities in my country, the installation rate of RFID equipment is still relatively low, and not all vehicles are equipped with RFID electronic tags, so the accuracy of travel time estimation and prediction based on RFID data is affected to a certain extent.

Images