Online learning method for optimizing signalized intersection queuing length

A technology of queuing length and learning method, which is applied in the field of online learning to optimize queuing length at signalized intersections, can solve problems such as inability to accumulate experience and form management plans, achieve real-time performance and adaptability, and improve computing efficiency

CN103077615AInactive Publication Date: 2013-05-01CHANGSHA UNIVERSITY OF SCIENCE AND TECHNOLOGY
0 Cites 38 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2013-05-01
Estimated Expiration
Not applicable · inactive patent

Smart Images

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

Abstract

The invention discloses an online learning method for optimizing a signalized intersection queuing length. The online learning method comprises the following steps of: 1, selecting states, behaviors and rewards; 2, reinforcing a learning matrix updating formula; 3, establishing a simulation optimizing platform; and 4, carrying out online operation. The online learning method is a signal timing dial optimizing technology which is capable of calculating a globally optimal solution and has the memorability. Compared with the risk neutral reinforcing learning technology, the online learning method has the advantages of no need of advanced offline learning, and better instantaneity and adaptability.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of traffic, and relates to an online learning method for optimizing the queuing length of a signalized intersection. Background technique

[0002] Artificial intelligence is an important direction in the field of modern science and technology, and an important means to realize system intelligence and improve system performance. Artificial intelligence methods are a cross-cutting technique that can be applied in various disciplines. In the field of road traffic signal control, the existing technology uses the science and technology from the 1960s to the 1980s, and these technologies have great limitations in dealing with huge state space and solution space problems. Signal timing at intersections of urban roads involves a huge state space and solution space, and the existing technology has made many simplifications to the problem of signal timing at intersections. Urban road network traffic congestion is a ...

Examples

Embodiment Construction

[0033] The technical solution of the present invention will be described in detail below in conjunction with specific examples of the accompanying drawings.

[0034]An online learning method for optimizing signal intersection queuing length, is characterized in that, comprises the following steps:

[0035] (1) Status, behavior, reward selection

[0036] 11) The vector composed of the queuing length of the key traffic flow in each phase is used as the state. In order to improve the calculation efficiency, the state space adopts a discrete form, and the discrete step length is an integer multiple of the average queuing length difference;

[0037] 12) The vector composed of the green light time of each phase is used as the behavior. For multi-phase intersections, the dimensionality disaster problem of behavior pairs will appear. The learning speed is the key to the practicality of online learning technology. In order to improve the learning speed, a dynamic behavior set is used. ...