Indoor Routing on Logical Network Using Space Semantics
College of Surveying and Geo-Informatics, Tongji University, Shanghai 200092, China
Faculty of Built Environment, the University of New South Wales, Sydney, NSW 2052, Australia
Department of OTB—Research for the Built Environment, Faculty of Architecture and the Built Environment, Delft University of Technology, 2628 BL Delft, The Netherlands
School of Urban Design, Wuhan University, Wuhan 430070, China
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2019, 8(3), 126; https://doi.org/10.3390/ijgi8030126
Received: 26 December 2018 / Revised: 20 February 2019 / Accepted: 23 February 2019 / Published: 3 March 2019
An indoor logical network qualitatively represents abstract relationships between indoor spaces, and it can be used for path computation. In this paper, we concentrate on the logical network that does not have notions for metrics. Instead, it relies on the semantics and properties of indoor spaces. A navigation path can be computed by deriving parameters from these semantics and minimizing them in routing algorithms. Although previous studies have adopted semantic approaches to build logical networks, routing methods are seldom elaborated. The main issue with such networks is to derive criteria for path computation using the semantics of spaces. Here, we present a routing mechanism that is based on a dedicated space classification and a set of routing criteria. The space classification reflects characteristics of spaces that are important for navigation, such as horizontal and vertical directions, doors and windows, etc. Six routing criteria are introduced, and they involve: (1) the spaces with the preferred semantics; and/or (2) their centrality in the logical network. Each criterion is encoded as the weights to the nodes or edges of the logical network by considering the semantics of spaces. Logical paths are derived by a traditional shortest-path algorithm that minimizes these weights. Depending on the building’s interior configuration, one criterion may result in several logical paths. Therefore, we introduce a priority ordering of criteria to support path selection and decrease the possible number of logical paths. We provide a proof-of-concept implementation for several buildings to demonstrate the usability of such a routing. The main benefit of this routing method is that it does not need geometric information to compute a path. The logical network can be created using verbal descriptions only, and this routing method can be applied to indoor spaces derived from any building subdivision.