基于CBAM注意力机制的智能交通信号控制

Intelligent Traffic Control Technology Based on CB AM Attention Mechanism

针对智能交通系统存在的卷积神经网络特征提取能力弱和特征表达能力有待提升等问题,在深度双Q网络(double deep Q network, Double DQN)模型基础上提出一种基于卷积注意力模块(convolutional block attention module, CBAM)的深度强化学习模型,用于智能交通信号控制。在三维卷积神经网络中加入CBAM轻量注意力模块,通过通道注意力和空间注意力两个模块结构更好地捕捉特征之间的相互依赖关系,增强卷积神经网络的特征表示质量,从而提升对拥堵路段重点特征的关注度以缓解交通拥堵问题。在城市交通仿真器SUMO(simulation of urban mobility)上的实验结果表明,相较其他常用算法,本文算法提高了交通灯配时的效率和稳定性,可为交通配时优化技术提供可靠依据。

A deep reinforcement learning model is pr oposed based on the double deep Q network(Double DQN)model,using the convolutional block at tention module(CBAM)to address issues such as weak feature extraction capabilities and li mited feature expression in convolutional neural networks within intelligent transportation system s.By integrating the lightweight CBAM attention module into the 3D convolutional neural network,th e model can better capture the interdependen-cies between features through the channel attentio n and spatial attention modules.This enhances the quality of feature representation in the convoluti onal neural network,thereby improving the focus on key features of congested road sections and allevia ting traffic congestion.Experimental results con-ducted on the SUMO(simulation of urban mobility)ur ban traffic simulator demonstrate that the proposed algorithm improves the efficiency and sta bility of traffic signal timing compared to other commonly used algorithms,providing a reliable basi s for traffic timing optimization technology.