Multi-intersection cooperative traffic light control method based on Q-value migration of multi-task deep Q network

A collaborative control, multi-intersection technology, applied in the intersection of artificial intelligence and intelligent transportation, can solve the problems of difficulty in extracting traffic state features, not fully mining the similarity of multiple agents, and lack of collaborative strategies for multi-intersection signal lights. The effect of alleviating congestion and improving traffic efficiency

CN110264750AActive Publication Date: 2019-09-20DALIAN UNIV OF TECH
6 Cites 24 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2019-09-20

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention, which belongs to the crossing field of artificial intelligence and intelligent transportation, provides a multi-intersection cooperative traffic light control method based on Q-value migration of a multi-task deep Q network. Multiple intersections are modeled into a multi-agent system; a multi-task network that is suitable for controlling of each intersection traffic light is trained at multiple intersections; knowledge learned at the multiple intersections is applied to a target problem, so that the multi-task network have much more knowledge than the single network and the single intersection feature extraction capacity is enhanced; and then on the basis of a cooperative algorithm, traffic lights of all intersections in a target area are controlled cooperatively. Therefore, the traffic flows of all intersections can be balanced to a certain degree; the road utilization rate in the regional traffic can be increased; the vehicle queuing length can be reduced; and the traffic congestion can be alleviated. The method has the high scalability to the traffic network.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the intersecting field of artificial intelligence and intelligent transportation, and relates to a multi-task deep Q-network-based Q-value transfer multi-intersection signal light cooperative control method. Background technique

[0002] Urban traffic runs through all areas of urban public space and is the most important and most convenient way for residents to travel daily. After long-term development, urban traffic has formed a relatively complete pattern, which improves the travel efficiency of residents and brings great benefits to residents' lives. convenient. However, with the development of the economy and the acceleration of urbanization, the urban population and the number of cars per capita are increasing rapidly, and the problem of traffic congestion is becoming more and more serious. . In order to alleviate traffic congestion, methods of expanding road infrastructure such as widening roads and building viaducts ca...

Examples

Embodiment Construction

[0049] The invention provides a multi-intersection signal lamp cooperative control method based on the Q value migration of the multi-task deep Q network. The specific embodiments discussed are merely illustrative of implementations of the invention, and do not limit the scope of the invention. Embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, specifically including the following steps:

[0050] 1. Obtain the status of the intersection. by figure 1 Take the six intersections in the example as an example. Each road has two lanes. According to the structure of the intersection, the left lane allows vehicles to go straight or turn left, and the right lane allows vehicles to go straight or turn right. Carry out discrete state encoding for the traffic information of each intersection, and divide the road k with length l from the stop line into discrete units of length c, where the value of c should be moderate, a...