The invention relates to a taxi scheduling visual analysis method and system based on multi-dimensional spatio-temporal data, in the technical scheme, rich visual channels are combined, different-dimension information of urban traffic data is displayed in sequence by using a multi-graph linkage mechanism, and resident travel rules and urban traffic system operation conditions implied in the data are explained. And tracking an urban traffic evolution rule under a multi-dimensional time-space view angle. The communication relation between urban hot spot areas and the radiation range of taxis are comprehensively considered, urban hot riding points and hot spot areas are accurately identified through a statistical algorithm and geometric knowledge, and the accuracy and reliability of urban area division are enhanced. On the basis, a multi-dimensional tensor is constructed, and a tensor decomposition algorithm is introduced to extract a potential resident travel mode in the tensor. Furthermore, the system integrates a real-time taxi demand prediction model based on deep learning, and combines the real-time travel demand with a resident travel mode analysis result, thereby improving the taxi scheduling efficiency and accuracy.
本发明涉及一种基于多维时空数据的出租车调度可视分析方法及系统,技术方案中结合丰富的视觉通道并利用多图联动机制依次展示城市交通数据的不同维度信息,解释隐含在数据下的居民出行规律和城市交通系统运行状况,以及追踪其在多维时空视角下的城市交通演化规律。综合考虑城市热点区域间连通关系以及出租车的辐射范围,通过统计学算法和几何学知识准确识别城市热门乘车点以及热点区域,增强城市区域划分的准确度和可靠性。在此基础上,构建多维张量并引入张量分解算法提取张量中的潜在居民出行模式。进一步地,系统集成基于深度学习的实时出租车需求预测模型,将实时出行需求与居民出行模式分析结果相结合,提升出租车调度效率和准确性。
Taxi scheduling visual analysis method and system based on multi-dimensional spatio-temporal data
基于多维时空数据的出租车调度可视分析方法及系统
2023-10-13
Patent
Electronic Resource
Chinese
IPC: | G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS |
Spatio-Temporal Influence of Extreme Weather on a Taxi Market
Transportation Research Record | 2021
|Optimizing Taxi Carpool Policies via Reinforcement Learning and Spatio-Temporal Mining
ArXiv | 2018
|