Reinforcement learning-based multi-agent system for network traffic signal control

Arel, Itamar; Liu, C.; Urbanik, Thomas; Kohls, Airton G

doi:10.1049/iet-its.2009.0070

Cited by 498 publications

(280 citation statements)

References 15 publications

Supporting

Mentioning

280

Contrasting

Order By: Relevance

“…It should be noted that we can also define e i,r (k) (or d i,r (k)), which obeys Poisson distribution [33,34]. In summary, in a practical application, we should use the historical data to estimate the distribution function of d i,r (k) and e i,r (k) reasonably.…”

Section: Uncertaintymentioning

confidence: 99%

Stochastic Model Predictive Control for Urban Traffic Networks

Gao

et al. 2017

Applied Sciences

View full text Add to dashboard Cite

This paper proposes a stochastic model predictive control (MPC) framework for traffic signal coordination and control in urban traffic networks. One of the important features of the proposed stochastic MPC model is that uncertain traffic demands and stochastic disturbances are taken into account. Aiming to effectively model the uncertainties and avoid queue spillback in traffic networks, we develop a stochastic expected value model with chance constraints for the objective function of the stochastic MPC model. The objective function is defined to minimize the queue length and the oscillation of green time between any two control steps. Furthermore, by embedding the stochastic simulation and neural networks into a genetic algorithm, we propose a hybrid intelligent algorithm to solve the stochastic MPC model. Finally, numerical results by means of simulation on a road network are presented, which illustrate the performance of the proposed approach.

show abstract

Section: Uncertaintymentioning

confidence: 99%

Stochastic Model Predictive Control for Urban Traffic Networks

Gao

et al. 2017

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Whereas Oliveira and Bazzan [13] propose an intersection control approach, based on agents who engage in coordination protocol, to collectively decide the direction of flow that should be prioritized. While Arel et al [14] introduced a reinforcement learning RL system multi-agent to obtain a control policy for efficient traffic lights, to minimize the average delay, congestion and probability of blocking in an intersection. Abdul Aziz et al [15] propose a technique called RMART, controlling the signal lights using the Markov decision process in a multi-agent framework.…”

Section: State Of the Artmentioning

confidence: 99%

Designing an Intelligent System for Traffic Management

Rezzai¹,

Dachry²,

Moutaouakkil³

et al. 2015

JCC

View full text Add to dashboard Cite

Abstract:Traffic congestion has become a serious problem that arises in our cities; this is due to rapid population growth, the rapid increase in the number of cars. Therefore, there is an infernal traffic jams. It is in this light, it is imperative to eliminate or at least reduce traffic congestion through the adoption of the policy of ITS (intelligent transport systems). The objective of this paper is to propose a study on the problem of congestion in cities, by reducing traffic congestion.

show abstract

“…Q-learning is one RL technique which is used in diverse applications, such as: multi-robot domains [2]; human-robot collaboration [9]; facade parsing [18]; robot navigation [5]; service discovery for wireless adhoc networks [7]; network traffic signal control [1]; nonplayer character decision making in video games [21]; computational modelling of electricity markets [12]; and dynamic pricing in electronic retail markets [6]. In these applications it is commonly assumed that the agent receives reinforcements (herein called rewards) instantaneously.…”

Section: Introductionmentioning

confidence: 99%

Multiple Model Q-Learning for Stochastic Asynchronous Rewards

Campbell

Givigi

Schwartz

2015

J Intell Robot Syst

View full text Add to dashboard Cite

This paper investigates reinforcement learning problems where a stochastic time delay is present in the reinforcement signal, but the delay is unknown to the learning agent. This work posits that the agent may receive individual reinforcements out of order, which is a relaxation of an important assumption in previous works from the literature. To that end, a stochastic time delay is introduced into a mobile robot line-following application. The main contribution of this work is to provide a novel stochastic approximation algorithm, which is an extension of Qlearning, for the time-delayed reinforcement problem. The paper includes a proof of convergence as well as grid world simulation results from MATLAB, results of line-following simulations within the Cyberbotics Webots mobile robot simulator, and finally, experimental results using an e-Puck mobile robot to follow a real track despite the presence of large, stochastic time delays in its reinforcement signal.

show abstract

Reinforcement learning-based multi-agent system for network traffic signal control

Cited by 498 publications

References 15 publications

Stochastic Model Predictive Control for Urban Traffic Networks

Stochastic Model Predictive Control for Urban Traffic Networks

Designing an Intelligent System for Traffic Management

Multiple Model Q-Learning for Stochastic Asynchronous Rewards

Contact Info

Product

Resources

About