基于修正IDM模型的交叉口跟驰模型仿真研究  

CAR-FOLLOWING MODEL OF SIGNALIZED INTERSECTION BASED ON MODIFIED IDM MODEL

作　　者：代宇婷 郭今达 宋利民[1] Dai Yuting;Guo Jinda;Song Limin(School of Information Science and Technology,Dalian Maritime University,Dalian 116026,Liaoning,China)

机构地区：[1]大连海事大学信息科学技术学院,辽宁大连116026

出　　处：《计算机应用与软件》2022年第9期72-79,共8页Computer Applications and Software

基　　金：国家自然科学基金项目(41671117)。

摘　　要：为了更精确地描述拥堵条件下的信号交叉口消散交通流行为,进而应用到智能交通并缓解城市交通拥堵,基于多个交叉口的高峰拥堵时段实测交通流数据,采用灰色关联分析法分析绿灯启亮时影响头车加速度的因素。在智能驾驶员模型(IDM)的基础上,构建一种适合交通拥堵时绿灯放行状态的基于改进头车启动模型的跟驰模型,该模型考虑了前方道路剩余长度对头车加速度的影响。利用遗传算法标定改进模型参数,对其进行效果验证,并与IDM模型进行仿真对比分析。仿真结果表明,该模型能很好地反映拥堵条件下绿灯启亮后的交叉口交通流行为,且明显优于重新标定的IDM模型,对解决智能交通基础问题和缓解城市交通拥堵具有一定的应用价值。In order to describe the dissipation traffic flow of signalized intersections under congestion conditions more accurately, and apply a model into intelligent traffic to relieve urban traffic congestion, on the basis of the measured traffic-flow data during peak congestion period of multiple intersections, we adopted gray correlation method to analyze the factors affecting the acceleration of head vehicles when the green light was on. Based on the intelligent driver model(IDM), we constructed a car-following model based on the improved first-car starting model, which was suitable for the green-light state at the intersection of traffic congestion. This model considered the influence of the remaining length of the road ahead on the acceleration of the first-car. Genetic algorithm was used to calibrate the parameters of the improved model. The effect of the model was verified and we compared the model with the IDM model. The simulation results show that the proposed model can reflect the actual traffic flow at the intersection when the green light is on under the congestion condition, and it is obviously better than the re-calibrate IDM model, which has certain application value for solving the basic problem of intelligent transportation and alleviating urban traffic congestion.

关 键 词：智能交通 交通流仿真 智能驾驶员模型 信号交叉口 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]