Travel time fusion prediction and query method based on traffic big data

Abstract

The invention discloses a travel time fusion prediction and query method based on traffic big data, and the method comprises the steps: carrying out the offline calculation or training of data uploaded by all online vehicles to obtain various types of prediction models and parameters, and building and dynamically updating a data dictionary according to the prediction models and parameters; carrying out the call of the prediction models and parameters, and carrying out the prediction of the traffic state of a road network or a single vehicle through combining the real-time travel time data of a road segment and a path, wherein the data dictionary comprises a vehicle data dictionary, a road segment data dictionary and a path data dictionary, and the online vehicles are the vehicles which are registered to access to the network and automatically upload positioning and speed data. The beneficial effects of the invention are that the method solves problems that a road network or single vehicle traffic state prediction method in the prior art is poorer in instantaneity, universality and practicality, greatly improves the measurement precision, greatly improves the online prediction precision, and guarantees the instantaneity, practicality and universality in engineering application.

Description

technical field [0001] The invention relates to road network or bicycle traffic state prediction technology, in particular to a travel time fusion prediction and query method based on traffic big data. Background technique [0002] The existing road network or bicycle traffic state prediction methods mainly use the traffic data provided by floating vehicles or roadside equipment, or even rely only on the traffic data provided by roadside equipment to predict the road network or bicycle traffic state. This type of method mainly obtains traffic information such as road network flow and average speed directly through roadside equipment, or obtains certain types of traffic information through secondary processing of floating car data, and uses model recursion to predict based on road network topology , the predicted traffic information is mainly concentrated on the three elements of traffic, resulting in a limited promotion area and low promotion value, that is, there is a gener...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems of poor real-time performance, poor versatility and poor practicability in the existing road network or bicycle traffic state prediction technology, this invention proposes a travel time fusion prediction and query method based on traffic big data

Images