You-Are-Here Maps as Location Based Service You-Are-Here Maps: Wayfinding Support as Location Based Service

Wayfinding support is a basic and important location based service. Today, this kind of service is mostly provided by solidly installed, static means. Of these means maps ideally support wayfinding, as it is possible, for example, to use one map for many visitors with many different destinations. A special kind of maps––You-Are-Here maps––is especially suited if design and placement criteria are carefully taken into account. In this paper we present a computational model that allows for determining those locations within an environment where maps function efficiently as location based wayfinding support. The model is based on findings in spatial cognition and, particularly, on research on You-Are-Here maps. The results of these fields of research have been analyzed to obtain operationable criteria for a prototypical system that calculates the placement of You-Are-Here maps for a given bounded outdoor environment. Even though the model is designed for ‘classical’ wayfinding support it already offers functionalities for current technological developments, i.e. electronic and mobile wayfinding systems. 1. WAYFINDING SUPPORT AS LOCATION BASED SERVICES In the light of new mobile information technologies like Personal Digital Assistants (PDAs), the Global Positioning System (GPS), and other mobile transmission standards it is getting constantly easier to provide location based services (LBS) to people acting in an environment. These services include, for example, information on interesting sights, location dependent news on traffic, weather or events, or advertisements for facilities located nearby. One of the most basic and at the same time most important services is supporting wayfinding processes. Orientation in an unknown environment is a critical factor for successfully arriving at a specified destination and requires usually external knowledge sources. Kai-Florian Richter & Alexander Klippel For car navigation commercial systems exist that help drivers find their way to a destination; for environments where people typically are on foot— like public buildings, convention centers, a university campus, or parks— these systems are still a matter of research (e.g., Baus et al., 2001; Malaka & Zipf, 2000). Officially provided wayfinding support in such environments, in opposition for instance to paper city maps, is usually solidly installed and static. Arguments for ‘Classical’ Wayfinding Support There are several technical and psychological reasons why solidly installed, static support is to prevail. From a technical perspective there are difficulties like providing everywhere a sufficiently stable connection between a transmitter of a localization signal and the receiving device; the costs to build an infrastructure that allows for such a service is rather high and, thus, needs to be re-financed via charges imposed on the users— something they do not accept yet. Furthermore, PDAs and other devices capable of services for mobile, electronic wayfinding support are not common enough by now. From a psychological perspective disadvantages of mobile electronic wayfinding support include, for instance, that the actual environment may be too complex to be adequately depicted on small displays PDAs or mobile phones offer. Among other things this hinders people gaining survey knowledge, which is the most complex kind of spatial knowledge and allows for the most inferences. Additionally, users may not want or be able to have their mobile devices turned on all the time and have to look at them for wayfinding information. Finally, as users may feel dependent on these devices, it reduces their acceptance. Maps as Wayfinding Support Two basic types of solidly installed, static wayfinding aids can be distinguished: maps and direction signs. While information provided by direction signs is normally faster to process than information represented in maps, these signs are problematic for the following reasons: signs show directions, not routes, i.e. at every decision point a new sign is needed; using a sign, just the direction to one destination can be shown resulting in an extra sign for every destination; signs are one way, i.e. following signs to one destination does not necessarily allow for finding the way back using the same signs; self localization and acquiring survey knowledge is 1 At least at every decision point that requires a direction change. You-Are-Here Maps as Location Based Service much more difficult when people depend solely on direction signs in wayfinding (cf O’Neill, 1999). Maps on the other hand are approved means for supporting wayfinding processes in an environment (MacEachren, 1986; Freksa, 1999): they allow to depict routes to different destinations, i.e. it is sufficient to provide one map instead of many signs at a given location; they show routes instead of directions, there is no need to provide a new map at every decision point; provided the map is sensibly positioned (cf. next section) self localization using a map is easy; maps ideally support acquiring survey knowledge (e.g. Freundschuh, 1991), hence, when using maps people do lose their dependence on wayfinding support earlier. On the other hand an analysis of maps used for wayfinding support reveals many design mistakes, as for the design of proper wayfinding aids a good understanding of the wayfinding process itself is necessary (cf. Arthur & Passini, 1992). Typically, this process consists of four sub-tasks: 1. Orientation, 2. Choosing the route, 3. Keeping the right track, 4. Discovering the objective (e.g., Downs & Stea, 1977; Daniel & Denis, 1998). Good wayfinding aids support all four of these tasks. Towards a Computational Model In this paper we outline a computational model for the identification of places at which maps function efficiently as location based wayfinding support. The model is based on findings in spatial cognition, especially research on cognitive maps (e.g., McNamara, 1991; Hirtle & Heidorn, 1993) and on landmarks (e.g., Sorrows & Hirtle, 1999), on constructive and interpretative processes on aspect maps (Berendt et al., 1998), and, in particular, on You-Are-Here maps (Levine, 1982). The next section reviews shortly research on YAH maps. We then present the computational model that allows for identifying locations important for wayfinding; the relevant criteria are introduced and it is shown how they can be combined in a model that is used to determine the locations. In section 4 we discuss an example; the model is applied to calculate locations for the placement of YAH maps on a university campus. Finally, we present some ideas on further research and possibilities to use the algorithm in mobile contexts. 2 There are signs that allow for self-localization, for example signs at the borders of states or cities or labels at buildings stating their number or function, e.g. ‘library’. Kai-Florian Richter & Alexander Klippel 2. YOU-ARE-HERE MAPS: A SHORT REVIEW As stated above maps are good means to support wayfinding processes. A special kind of maps used is You-Are-Here (YAH) maps (Levine, 1982; O’Neill, 1999). These maps are solidly installed in the environment–– usually vertically attached on walls or signposts––and contain a You-AreHere symbol. This symbol indicates on the map the location where in the environment the map is positioned. The symbol significantly eases selflocalization as it obviously indicates ones position within an environment. Levine (1982) identified design and positioning criteria to enhance a YAH map’s suitability for aiding wayfinding. O’Neill (1999) resumed the original work and extended the requirements for YAH maps resulting in the following list: • Provision of signs and labels in the environment and correspondingly on the map: this eases the mapping between what is seen on the map and what is seen in the environment itself. If signs and labels seen on the map can be detected in the environment, relating features of the map to features of the environment is straightforward. Thus, self-localization as well as orientation and choosing the route become much simpler. • Inclusion of architectural cues and natural landmarks: YAH maps should be designed such that architectural cues and natural landmarks are included and that the shape of the paths drawn in the YAH map relates to the actual shape of the paths found in the environment. As these are important cues in people’s wayfinding behavior this eases map use and reinforces learning the environment’s layout. • Placement of YAH maps near entrances of an environment and at decision points: these are the locations in an environment where people (re)orient themselves and, thus, need support for their decisions. • Placement of YAH maps near asymmetrical parts of an environment: this facilitates to locate the map within the environment. An asymmetrical part of an environment is easily identified on the map as its layout combined with the YAH symbol shown on the map provides many cues for its location. Therefore, the location of the map in the environment becomes non-ambiguous. • Use of a complex YAH symbol: a complex YAH symbol shows, along with the location of a map in the environment, an orientation, which is the viewing direction of the user. This significantly eases orientation at the corresponding location and determining the path to the destination as, again, relating map features to features in the environment is You-Are-Here Maps as Location Based Service facilitated; alignment (see next paragraph) and the orientation of the complex YAH symbol should correspond to each other. • Alignment of the map and environment: if a map is aligned with the environment the relation top in the map corresponds to the relation in front of of the user viewing the map. Hence, a corresponding relative reference system is established in which the relations ‘left’ and ‘right’ are the same in the map and in the viewed environment; this greatly helps mapping what is seen on the map to what is seen in the environment (see also Shepard & Hurwitz, 1984). • Redundancy: combining the principles mentioned above allows for easy self-localization, orientation and determination of the path to the destination (see also Hirtle, 2000). Especially map alignment has been subject to further research (e.g. Shepard & Hurwitz, 1984; Warren & Scott, 1993). Apart from the YAH symbol this is the most important criterion for the construction of good YAH maps; people usually expect YAH maps to be aligned and they use them accordingly (Levine et al., 1984). Thus, non-aligned YAH maps significantly complicate the wayfinding process. Still many YAH maps found in environments are often not adequately designed. Sometimes the YAH symbol is missing and, often, these maps are not properly aligned. Figure 1 shows a typical example: this map can be found on the island Wangerooge in the northwest of Germany. While a YAH symbol is provided and, thus, simple perceptual processes accomplish self-localization, the map is not aligned; determining a route to take is, therefore, unnecessarily aggravated. 3 This is the main reason why people tend to rotate maps they hold in their hands when using them for