Network-wide traffic scheduling optimization method, device, computer equipment and storage medium

Abstract

The present invention relates to the field of artificial intelligence technology, and provides a network-wide traffic scheduling optimization method, device, computer equipment, and storage medium, including: training and building a distributed GPU scheduling cluster based on a gradient descent algorithm to obtain a distributed traffic identification model, each The GPU corresponds to a traffic intersection; obtain the environmental image collected by the image acquisition device; distribute the environmental image to the GPU corresponding to the traffic intersection according to the identification information of the traffic intersection; identify the environmental image through the traffic recognition model in the GPU to obtain the traffic congestion degree of the traffic intersection ; According to the traffic congestion degree of all traffic intersections and the corresponding traffic intersection signs, the traffic is dispatched and optimized. The invention can be applied to intelligent transportation, can quickly identify the traffic congestion degree of all traffic intersections at the same time, and can dispatch and optimize the traffic of the whole network. In addition, the present invention also relates to block chain technology, and the distributed traffic identification model can be stored in the block chain.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a method, device, computer equipment and storage medium for optimizing traffic dispatching across the network. Background technique [0002] The continuous development of modern transportation has brought great convenience to people's lives, but various traffic problems have also occurred frequently. When vehicles are running on the road, traffic congestion, road blockage, and traffic accidents occur frequently, which greatly affects travel time and traffic efficiency. [0003] In the prior art, by monitoring the traffic flow image at the intersection, combined with the neural network model to identify the traffic flow status at the monitored intersection, determine whether traffic congestion occurs at the monitored intersection, and then perform scheduling optimization. However, this monitoring method can only realize the scheduling optimization of a sin...

AI Technical Summary

Problems solved by technology

However, this monitoring method can only realize the scheduling optimization of a single intersection, and cannot realize the overall scheduling optimization of the comprehensive road network.

Images