摘要
针对一阶马尔可夫链在不确定性轨迹预测中预测准确率不高的问题,本文提出构建基于二阶马尔可夫链的不确定性轨迹预测模型。以北京市出租车轨迹数据为例,分别构建模型预测2 min、5 min、10 min内出租车的位置,并对预测结果准确率进行分析对比。研究结果表明,基于二阶马尔可夫链比基于一阶马尔可夫链的不确定性轨迹预测模型的预测准确率提高了40%左右;模型预测准确率随预测时间的延长而下降;模型预测结果较好,具有一定的应用价值。
Aiming at the problem that the lower accuracy of the first-order Markov chain in the uncertainty trajectory prediction,this paper proposes to construct an uncertainty trajectory prediction model based on second-order Markov chain.Taking the taxi trajectory data of Beijing as an example,the model is constructed to predict the location of taxis within 2 min,within 5 min,and within 10 min,and analyzed the accuracy of prediction results.The results show that the prediction accuracy of the uncertainty trajectory prediction model based on the second-order Markov chain is improved by about 40%.The prediction accuracy of the model decreases with the prediction time;Model prediction The result is good and has certain application value.
作者
冯然
张力仁
王立辉
FENG Ran;ZHANG Liren;WANG Lihui(The Second Surveying and Mapping Engineering Institute of Heilongjiang,Harbin 150025,China;The Fifth Surveying,Mapping and Geographic Information Engineering Institute of Heilongjiang Province,Harbin 150081,China)
出处
《测绘与空间地理信息》
2020年第S01期207-211,共5页
Geomatics & Spatial Information Technology
关键词
出租车轨迹数据
二阶马尔可夫链
轨迹预测
taxi trajectory data
second-order Markov chain
trajectory prediction
