We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AI

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Artificial Intelligence

Title: Network-wide traffic signal control optimization using a multi-agent deep reinforcement learning

Abstract: Inefficient traffic control may cause numerous problems such as traffic congestion and energy waste. This paper proposes a novel multi-agent reinforcement learning method, named KS-DDPG (Knowledge Sharing Deep Deterministic Policy Gradient) to achieve optimal control by enhancing the cooperation between traffic signals. By introducing the knowledge-sharing enabled communication protocol, each agent can access to the collective representation of the traffic environment collected by all agents. The proposed method is evaluated through two experiments respectively using synthetic and real-world datasets. The comparison with state-of-the-art reinforcement learning-based and conventional transportation methods demonstrate the proposed KS-DDPG has significant efficiency in controlling large-scale transportation networks and coping with fluctuations in traffic flow. In addition, the introduced communication mechanism has also been proven to speed up the convergence of the model without significantly increasing the computational burden.
Subjects: Artificial Intelligence (cs.AI)
Journal reference: Transportation Research Part C: Emerging Technologies Volume 125, April 2021, 103059
DOI: 10.1016/j.trc.2021.103059
Cite as: arXiv:2104.09936 [cs.AI]
  (or arXiv:2104.09936v1 [cs.AI] for this version)

Submission history

From: Zhenning Li [view email]
[v1] Tue, 20 Apr 2021 12:53:08 GMT (2389kb,D)

Link back to: arXiv, form interface, contact.