A novel cooperative MARL-based approach for adaptive traffic signal control in multi-intersection networks that significantly outperforms existing methods in relation to average pheromone intensity, average noise emission, and average waiting time is proposed.
This study proposes a distributed traffic signal control framework built upon a Machine Learning (ML) paradigm utilizing Reinforcement Learning (RL), and demonstrates the effectiveness of the proposed approach, with vehicle queue lengths and average waiting times reduced by 35% on roads leading to the junctions, compar...
This paper targets traffic signal control in Japan and proposes an implementable signal control method by combining reinforcement learning with traditional traffic engineering, demonstrating the potential of integrating traffic engineering with RL to develop effective signal control methods.
A demand-driven signal control strategy is developed to allocate green time based on real-time vehicle demand, eliminating wasted signal phases and providing a scalable and intelligent solution for modern smart city traffic systems.
A hybrid reinforcement learning approach for traffic signal control that combines the complementary learning mechanisms of Q-learning, SARSA, and Monte Carlo algorithms to improve both learning efficiency and control performance is proposed.
SelectLight is proposed, which implements post-optimization selection by allowing a multi-agent reinforcement learning (MARL) policy to choose directly from plans generated online by DMPC, and achieves the best delay-related performance and that its advantage widens with demand.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
