What question did this study set out to answer?

To develop and evaluate a deep reinforcement learning framework that simultaneously optimizes traffic signal phase sequence and timing to enhance intersection efficiency.

January 1, 2025Digital Transportation and SafetyOpen Access

Joint control of traffic signal phase sequence and timing: a deep reinforcement learning method

View Full Paper

Ask AI

Bookmark

Authors

ZSZhanbo SunSouthwest Jiaotong University YCYiming CaiJilin University AJAng JiSouthwest Jiaotong University

Discussion

Member takes

Implication

Simulation study reveals joint reinforcement learning control of signal phase and timing cuts queue lengths by over 7.56%, indicating superior traffic efficiency over traditional baselines.

Key Points

To develop and evaluate a deep reinforcement learning framework that simultaneously optimizes traffic signal phase sequence and timing to enhance intersection efficiency.
Designed a 3DQN framework incorporating prioritized experience replay to jointly manage phase sequence transitions and green-light durations.
Integrated macroscopic feature-based and microscopic cell-based traffic state representations to guide agent decision-making.
Evaluated performance against conventional Webster and MaxPressure baselines, as well as isolated phase-switching and phase-duration models in traffic simulations.
The proposed 3DQN method with prioritized experience replay outperformed traditional baselines, reducing intersection queue lengths by at least 7.56%.
Relying on microscopic information alone increased queue lengths by 2.44% relative to combined macro-micro states, while macroscopic representation alone failed to achieve training convergence.
Joint optimization of sequence and duration achieved a 6.37% greater queue length reduction than phase-switching control alone.

Cite This Study

Sun et al. (2025) studied this question.

synapsesocial.com/papers/6a7d4d92ff1a536db837d78c https://doi.org/10.48130/dts-0025-0008

View Full Paper

Ask AI

Bookmark

January 1, 2025Digital Transportation and SafetyOpen Access

Joint control of traffic signal phase sequence and timing: a deep reinforcement learning method

View Full Paper

Ask AI

Bookmark

Authors

ZSZhanbo SunSouthwest Jiaotong University YCYiming CaiJilin University AJAng JiSouthwest Jiaotong University

Discussion

Member takes

Implication

Simulation study reveals joint reinforcement learning control of signal phase and timing cuts queue lengths by over 7.56%, indicating superior traffic efficiency over traditional baselines.

Key Points

To develop and evaluate a deep reinforcement learning framework that simultaneously optimizes traffic signal phase sequence and timing to enhance intersection efficiency.
Designed a 3DQN framework incorporating prioritized experience replay to jointly manage phase sequence transitions and green-light durations.
Integrated macroscopic feature-based and microscopic cell-based traffic state representations to guide agent decision-making.
Evaluated performance against conventional Webster and MaxPressure baselines, as well as isolated phase-switching and phase-duration models in traffic simulations.
The proposed 3DQN method with prioritized experience replay outperformed traditional baselines, reducing intersection queue lengths by at least 7.56%.
Relying on microscopic information alone increased queue lengths by 2.44% relative to combined macro-micro states, while macroscopic representation alone failed to achieve training convergence.
Joint optimization of sequence and duration achieved a 6.37% greater queue length reduction than phase-switching control alone.

Cite This Study

Sun et al. (2025) studied this question.

synapsesocial.com/papers/6a7d4d92ff1a536db837d78c https://doi.org/10.48130/dts-0025-0008

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark