基于混沌遗传算法的交通灯时长调节仿真  

Simulation of Traffic Lamp Duration Regulation Based on Chaotic Genetic Algorithm

作　　者：杨道平 葛耿育 夏德友 YANG Dao-ping;GE Geng-yu;XIA De-you(School of Information Engineering,Zunyi Normal College,Zunyi Guizhou 563006,China;School of Computer Science and Technology,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]遵义师范学院信息工程学院,贵州遵义563006 [2]重庆邮电大学计算机科学与技术学院,重庆400065

出　　处：《计算机仿真》2023年第5期192-196,共5页Computer Simulation

基　　金：贵州省科技基金(黔科合基础[2018]1180号)。

摘　　要：路口交通信号灯的时长受车辆实时流量和通行距离约束,且车辆的不同去向也会引起差异较大的时间差,因此当前交通灯仍无法根据实际交通情况的自适应调节时长。为此,提出基于混沌遗传算法的交通灯时长自适应调节方法。通过对车辆、道路网络、行驶状态的描述,构建交通控制模型。基于此,采用改进混沌遗传算法优化模型参数。通过模拟退火算法加快混沌遗传算法的种群进化效率,解码最优个体,获得最优参数值,求解最优交通灯时长自适应调节结果。实验结果表明,所提方法在车辆数为2000辆时吞吐量为1.8veh·s-1,车辆等候时间均值为35s,车辆饱和度在1600辆时始终低于0.6;滞留车辆数目更少,车辆总数达到100时滞留车辆仅在10辆左右。The duration of traffic lights at intersections is constrained by the real-time traffic and travel distance of vehicles.However,it is unable to adjust the duration of traffic lights adaptively according to actual traffic condi-tions.Therefore,this paper presented an adaptive adjustment method for traffic lights based on chaotic genetic algo-rithm.Based on the description of vehicles,road network and driving state,a traffic control model was constructed.On this basis,the improved chaotic genetic algorithm was used to optimize the parameters of the model.Then,the simulated annealing algorithm was used to accelerate the population evolution efficiency of a chaotic genetic algorithm and decode the optimal individual,thus obtaining the optimal parameter value.Finally,the adaptive adjustment of the optimal traffic light duration was solved.Experimental results show that the throughput of the proposed method is 1.8veh·S^(-1) when the number of vehicles is 2000.The average waiting time of vehicles is 35s,and the saturation of vehicles is always less than 0.6 when the number of vehicles is 1600.The number of stranded vehicles is 10 when the total number of vehicles is 100.

关 键 词：混沌算法 遗传算法 模拟退火算法 交通灯时长 自适应调节 

分 类 号：TP391.44[自动化与计算机技术—计算机应用技术] U495[自动化与计算机技术—计算机科学与技术]