Method and device for training scheduling model, and method and device for realizing collaborative driving

Abstract

The invention discloses a method and a device for training a scheduling model, and a method and a device for realizing cooperative driving, and the method comprises the steps: carrying out the embedding processing of sample vehicle state information, and mapping a corresponding high-dimensional state vector of each first vehicle; processing the obtained high-dimensional state vector to obtain association relationship information corresponding to each first vehicle; determining passing sequence information of the first vehicle according to the obtained association relationship information; calculating the sum of delays of all the first vehicles passing through the non-signalized intersection according to the obtained passing sequence information; and determining a deep learning model for vehicle scheduling according to the calculated delay sum. According to the embodiment of the invention, the scheduling model which can be used for vehicle scheduling is obtained through offline sample data training, the vehicle scheduling time is shortened through the scheduling model, and the vehicle scheduling efficiency and quality are improved.

Description

Method and device for training scheduling model, and method and device for realizing collaborative driving technical field This paper relates to but not limited to neural network technology, especially refers to a kind of method, device, realization coordination of training scheduling model Method and device for driving. Background technique [0002] The intelligent vehicle-road coordination system adopts advanced wireless communication and fast edge computing and other technologies to realize the comprehensive realization of vehicle Vehicle-to-vehicle, vehicle-to-roadside equipment information sharing. The vehicle cooperative driving technology is based on the collected real-time traffic and Vehicle information, using a centralized decision-making and control method, can not only ensure the traffic safety in the process of vehicle driving, but also It can significantly improve the efficiency of the transportation system and is a brand-new technical route to realize a...

AI Technical Summary

Problems solved by technology

Since the right-of-way assignment problem at an intersection is a non-deterministic polynomial (NP)-hard problem, the methods provided by related technologies can only deal with scenes with fewer vehicles (less than 15 vehicles), and for more general scenarios with more vehicles scenarios (more than 20 vehicles), it is impossible to provide a real-time and efficient decision-making scheme for road right allocation; in addition, most of the methods of related technologies rely on the topological structure of the intersection. It cannot be deployed efficiently and quickly to the urban road network traffic system with various intersections

Images