Indoor Signposting and Wayfinding through an Adaptation of the Dutch Cyclist Junction Network System

Abstract

Finding one's way in complex indoor settings can be a quite stressful and time-consuming task, especially for users unfamiliar with the environment. There have been several different approaches to provide wayfinding assistance in order to guide a person from a starting point to a destination, ranging from traditional maps and signposting systems to the most recent developments of navigation systems displaying direction information on a mobile device screen. However, so far none of them has emerged to be efficient enough in order to act as a uniform solution for the wayfinding problem in indoor environments. Moreover, referencing to landmarks is not widely employed by wayfinding assistance systems despite the fact that landmark-based navigation is the most common way for people to navigate themselves through unfamiliar environments. The aim of this work is to propose a new wayfinding assistance method for indoor environments that makes use of the landmark concept. The method to achieve it is by translating the main principles of an already existing outdoor system named Junction Network System, which applies successfully for providing guidance to cyclists in The Netherlands. The most important parameter of the system is that assistance is provided to users where they actually need to make a directional choice. The first step is to automatically define the locations, i.e. the decision points, where wayfinding assistance is needed in indoor settings and the second is to supply them with a special type of landmark which will have the form of a signpost that provides all the necessary information. A graph based representation of the indoor setting is generated in order to extract the decision points and create the network of all possible routes in the environment. The conceptual framework for the determination of the decision points is implemented in two case studies: GeoFort main building and the Faculty of Architecture and the Built Environment (BK) of TU Delft. Two different building cases are used in order to gain an insight on how the system can be applied in different indoor environments. The part of the navigable indoor space where movement mainly occurs when people are in search of a destination was subdivided according to a geometric criterion (Constrained Delaunay Triangulation) while also the semantics of the space were taken into account in order to determine the decision points. By connecting all the decision points and the destinations a network of all the possible routes is created. A route is followed by referring to the sequence of numbers corresponding to the nodes of the network. The proposed system was verified based on a human-based survey. The proposed landmark-signs were installed at part of the faculty of Architecture and users were asked to find their way in the building relying only on them. The time to reach the destination and the number of detours were counted, while a questionnaire-based survey was used in order to further investigate the feelings and views of participants regarding the effectiveness of the system. The findings of the experiment are quite promising, as the performance of the participants in both measures was very good. Additionally, the results of the questionnaire indicate that users consider the system simple and comprehensive and they are in general satisfied by the provided assistance. However, further implementation and testing in more indoor space cases is needed in order to draw an indisputable conclusion about its effectiveness. Nevertheless, an important contribution of the research is the generation of a conceptual model in order to automatically determine the decision points in indoor environments that this can be applied invariable to other wayfinding assistance systems.