Abstract: In order to make up for the shortcomings of the traditional path navigation service within the built environment, an optimal path analysis algorithm of indoor and outdoor integration is proposed. To achieve this goal, a 2D geometric network model is introduced, and a coordinate transform strategy of indoor spatial data is designed to realize the rapid construction and visual expression of the indoor 3D route topology data model. The Dijkstra shortest path algorithm provided by the open source pgRouting library is extended to realize the optimal path and corresponding direction information query between any input two points based on PostgreSQL. Finally, an indoor and outdoor integrated optimal route query prototype system has been developed on the basis of open source software, such as GeoServer and OpenLayers. The performance of the proposed approach is tested on a large-scale indoor and outdoor integrated route network, and the qualitative and quantitative analysis results demonstrate that this method is valid and efficient. The proposed approach is compatible with the common urban traffic network data and mature shortest path analysis algorithms, which is universal and practical.

Key words: indoor and outdoor integration, optimal path, geometric network model, pgRouting, Dijkstra

摘要： 为了弥补传统路径导航服务在室内立体空间方面的不足，提出了一种室内外一体化的网络数据模型和最优路径分析解决方案。以几何网络模型为基础，设计了一种楼层数据偏移策略，实现室内三维空间路径拓扑模型快速构建和二维可视化表达。对开源pgRouting库内置的高效Dijkstra路径查询函数进行扩展，实现了基于PostgreSQL库的任意两点之间最优路径和转弯方向语义信息查询。最后，利用GeoServer和OpenLayers等开源软件开发了室内外一体化路径查询原型系统，并采用大规模室内外一体化路径网络模型数据进行测试，定性与定量分析对比结果验证了该方法的正确性和高效性。该方法能够最大化兼容城市交通网络数据和成熟的最短路径分析算法，具有普适性与实用性。

关键词: 室内外一体化, 最优路径, 几何网络模型, pgRouting, Dijkstra