Traffic Signal Optimization Control Method Based on Bayesian Deep Q Network

By optimizing traffic signal control using Bayesian deep Q-networks, the inefficiency of traditional methods in complex traffic scenarios is solved, achieving efficient dynamic adjustment of traffic flow and energy conservation.

CN116824848BActive Publication Date: 2026-07-17GANSU ZIGUANG INTELLIGENT TRANSPORTATION & CONTROL TECH
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
GANSU ZIGUANG INTELLIGENT TRANSPORTATION & CONTROL TECH
Filing Date
2023-06-08
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Traditional traffic signal control methods are ill-suited to handling time-varying and complex large-scale traffic scenarios, leading to traffic congestion, longer travel times, and energy waste.

Method used

A reinforcement learning method based on Bayesian deep Q-networks is adopted to construct a traffic signal control model. By utilizing the reinforcement learning capabilities of Bayesian deep Q-networks, the traffic signal control strategy is optimized. Combined with real-time traffic data and simulation models, dynamic signal adjustment is achieved.

Benefits of technology

It improves traffic flow efficiency, reduces traffic delays and energy consumption, enhances road capacity, strengthens the robustness and adaptability of the model, and enables it to respond promptly to changes in traffic flow.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116824848B_ABST
    Figure CN116824848B_ABST
Patent Text Reader

Abstract

The traffic signal optimization control method based on Bayesian deep Q-networks, relating to the field of intelligent transportation technology, includes the following steps: (1) establishing a traffic signal control model based on deep reinforcement learning, specifically including: s11 defining the state of the model; s12 defining the actions of the model; s13 defining the reward function of the model; s14 designing a priority Bayesian deep Q-network; (2) training a multi-intersection collaborative adaptive signal timing optimization control model based on deep reinforcement learning; (3) generating a traffic signal timing optimization control strategy and continuously updating the model. This method utilizes the reinforcement learning capability of Bayesian deep Q-networks to optimize traffic signal control, thereby improving traffic flow efficiency and reducing traffic congestion.
Need to check novelty before this filing date? Find Prior Art