基于神经网络的交通灯智能决策系统设计  

Design of Traffic Light Intelligent Decision System Based on Neural Network

作　　者：金晓磊 陈苗 刘喜庆 JIN Xiaolei;CHEN Miao;LIU Xiqing(College of Electronics and Information,Xi'an Polytechnic University,Xi'an 710600,China;Wujiaqu Power Supply Company,Xinjiang Electric Power Company of State Grid,Wujiaqu Xinjiang 831300,China)

机构地区：[1]西安工程大学电子信息学院,西安710600 [2]国网新疆电力公司五家渠市供电公司,新疆五家渠831300

出　　处：《微处理机》2020年第1期57-60,共4页Microprocessors

摘　　要：为缓解城市交通拥堵,提高交通灯智能化水平,提出基于神经网络的交通灯智能决策系统设计。设计依据全面性原则、明确性原则、实用性原则选取车速、车流量和交通密度作为决策指标;创建了一个三输入、一输出的三层BP神经网络模型,用搜集的实际交通数据进行训练,并对训练后得到的模型的有效性进行验证;设计了交通灯决策系统GUI,根据当前采集到的路况信息,利用该神经网络模型决策交通灯的适宜绿灯时间。经仿真验证,利用BP神经网络根据当前交通状况可以决策出一个较为适宜的交通灯绿灯时间。In order to alleviate traffic congestion and improve the intelligence level of traffic lights,an intelligent decision-making system for traffic lights based on neural network is proposed.The design is based on the principles of comprehensiveness,clarity and Practicability to select speed,traffic flow and traffic density as decision indicators.A three-layer BP neural network model with three inputs and one output is established,and the collected actual traffic data are used for training,and the effectiveness of the trained model is verified.A traffic light decision system GUI is designed.According to the collected traffic information,the neural network model is used to decide the appropriate green time of the traffic light.The simulation results show that the BP neural network can be used to decide a more suitable traffic light green time according to the current traffic situation.

关 键 词：BP神经网络 智能交通灯 智能决策系统 图形用户界面 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]