Urban vehicle-road cooperative traffic flow prediction method based on digital twinning

Abstract

The invention discloses an urban vehicle-road cooperation traffic flow prediction method based on digital twinning, is applied to the field of Internet of Vehicles, and aims to solve the problems that large-range overall prediction of an urban system cannot be realized and high precision cannot be guaranteed at the same time in the current traffic flow prediction research under the limitation of computing power, and a model is difficult to adjust in real time in case of emergencies. Layered twinning is established in a virtual space for the overall topology of a city, the internal topology of each region and vehicle individuals, so that spatial-temporal characteristics of complex and large-range urban traffic flow are excavated. Cooperative prediction among different areas is realized by dividing urban areas and establishing overall area topology, so that the problem of computing power burden caused by a centralized mode is solved; through a deduction process among layers in digital twinning, traffic flow prediction and traffic guidance decision are continuously interacted, periodic feedback is generated for vehicles, and the vehicles upload perception data in real time, so that the overall prediction precision is improved.

Description

technical field [0001] The invention belongs to the field of intelligent vehicle networking, in particular to an urban traffic flow prediction technology. Background technique [0002] Traffic flow prediction is an important basis for the realization of intelligent transportation systems. With rapid urbanization and population growth, intelligent transportation systems are becoming more complex, and people's demands for travel safety and efficiency are also increasing. Early intervention based on traffic prediction and traffic scheduling are the keys to improving road efficiency, driving safety and alleviating traffic congestion in urban transportation systems. In recent years, with the support of 5G technology and the sensor communication computing function of intelligent vehicles, the data communication between the vehicle and the infrastructure network has been realized through the 5G vehicular ad hoc network (5G-VANET), which provides a solution for real-time traffic fl...

AI Technical Summary

Problems solved by technology

The classic methods include DCRNN, ASTGCN, etc. However, the above schemes have the following disadvantages: 1. Most of them only consider roads or regions, and do not consider traffic flow prediction at the city scale

2. The prediction of the future based on the historical database does not consider the solution of real-time road data at the same time, which will also lead to the problem of insufficient prediction accuracy

After searching the existing literature, most of the relevant attempts in the intersection of digital twins and transportation systems only start from the individual and do not consider the system level

Images