DYNAMIC PREDICTION OF TRAFFIC VOLUME THROUGH KALMAN FILTERING THEORY
DYNAMIC PREDICTION OF TRAFFIC VOLUME THROUGH KALMAN FILTERING THEORY
复制标题
DOI：
10.1016/0191-2615(84)90002-x
复制
发表时间：
1984-01-01
影响因子：
6.8
通讯作者：
STEPHANEDES, YJ
中科院分区：
文献类型：
--
作者：
OKUTANI, I;STEPHANEDES, YJ
提出了两种基于卡尔曼滤波理论的短期交通量预测模型。使用最近的预测误差来改进预测参数，并且通过考虑来自多个链路的数据来实现对链路的更好的流量预测。基于名古屋市某道路网的数据，平均预测误差小于9%，最大误差小于30%。新型号的性能比UTCS-2好很多(高达80%)。
Two models employing Kalman filtering theory are proposed for predicting short-term traffic volume. Prediction parameters are improved using the most recent prediction error and better volume prediction on a link is achieved by taking into account data from a number of links. Based on data collected from a street network in Nagoya City, average prediction error is found to be less than 9% and maximum error less than 30%. The new models perform substantially (up to 80%) better than UTCS-2.