基于ANFIS的可变导向车道智能控制系统

发布时间：2018-03-01 08:55

本文关键词： 交通控制交叉口可变导向车道交通状态控制策略自适应模糊神经推理系统　出处：《交通运输工程学报》2017年04期 　论文类型：期刊论文

【摘要】：为缓解因交通流向分布不均衡导致的交叉口交通拥挤状况,以交叉口进口道的可变导向车道为研究对象,建立了基于自适应模糊神经推理系统的可变导向车道智能控制系统。智能控制系统由数据采集子系统、交通状态预测子系统和控制子系统构成,共同完成可变导向车道的智能化控制。将数据采集子系统检测到的实时交通数据录入到预先训练好的交通状态预测子系统中,可得到左转车辆和直行车辆的运行状态,并根据控制子系统的结构化算法来确定可变导向车道的属性。计算结果表明:交通状态预测子系统的测试误差为0.075 097,满足精度要求,可以用于交通状态预测;采用可变导向车道智能控制系统能明显改善交叉口交通拥堵状况,当左转车辆比例为25%时,关键进口道综合延误减少了6.1%,平均停车次数减少了9.5%,平均排队长度减少了6.1%,当左转车辆比例上升至30%时,3个指标分别下降了8.1%、12.4%与8.0%,表明左转比例越高,作用效果越显著。
[Abstract]:In order to alleviate the traffic congestion at intersections caused by the uneven distribution of traffic flow direction, the variable steering lane of intersections entrance is taken as the research object. The intelligent control system of variable steering lane based on adaptive fuzzy neural inference system is established. The intelligent control system is composed of data acquisition subsystem, traffic state prediction subsystem and control subsystem. The real-time traffic data detected by the data acquisition subsystem is input into the pre-trained traffic state prediction subsystem, and the running state of the left-turn vehicle and the direct vehicle can be obtained. According to the structured algorithm of the control subsystem, the attributes of the variable steering lane are determined. The results show that the test error of the traffic state prediction subsystem is 0.075,097, which meets the precision requirement and can be used for traffic state prediction. The intelligent control system of variable steering lane can obviously improve the traffic congestion at the intersection, when the ratio of left-turn vehicles is 25, The comprehensive delay of the key entrance is reduced by 6.1%, the average number of stops decreases by 9.5%, and the average queue length decreases by 6.1%. When the proportion of left-turn vehicles rises to 30%, the three indexes decrease by 8.1% and 8.0%, respectively, indicating that the higher the left-turn ratio, the more significant the effect.
【作者单位】：西南交通大学交通运输与物流学院;西南交通大学综合交通运输智能化国家地方联合工程实验室;
【基金】：国家自然科学基金项目(51774241,51675451)
【分类号】：U491

【相似文献】