Expressway road condition identification and prediction method based on floating vehicle data

Abstract

The invention discloses an expressway road condition identification and prediction method based on floating vehicle data, and the method comprises the following steps: carrying out the data preprocessing based on the original GPS track data, based on a speed calculation method for grid division results, calculating a travel vehicle speed by using an Euclidean distance between track points and travel time as an instantaneous speed of a vehicle at an updating point; based on the traffic flow parameters in the speed calculation grid, calculating the traffic flow parameter value in each unit by taking the space grid and the time grid as units; dimensionally reducing traffic state parameters based on principal component analysis, simiplifying the dimension of data and abandoning irrelevant features in the data; based on k-means traffic state clustering anylsis, identifying different traffic states; and constructing characteristics of different time scales, and establishing each traffic state quantity prediction model based on the long short-term memory neural network LSTM. The expressway traffic state can be accurately identified, and the evolution trend of the expressway traffic statecan be predicted.

Description

technical field [0001] The invention relates to the technical field of traffic big data, in particular to a method for identifying and predicting highway traffic conditions based on floating car data. Background technique [0002] With the continuous and rapid development of social economy, the expressway network, as the national artery, plays an important role in the national economic construction and development, and maintaining its good operation status has become an important figure in the traffic management department. In the past, the monitoring and prediction of highway status mainly relied on fixed detector data, which often has the characteristics of limited coverage and inflexibility. [0003] With the promotion of GPS equipment, the magnitude of floating car data continues to increase; compared with the high maintenance and management costs of traditional detectors, floating car data has the characteristics of simple acquisition and low cost; therefore, floating c...

AI Technical Summary

Problems solved by technology

In the past, the monitoring and prediction of highway status mainly relied on fixed detector data, which often has the characteristics of limited coverage and inflexibility

Images