As described on our Research Interests page, we have a particular interest in indoor navigation because it involves a number of challenges, both spatiocognitive and technological, which differentiate it from navigation in other types of environments. The crux of the issue is that compared to outdoor travel, indoor navigation is supported by far less environmental information, orienting cues, and external aids (such as maps or GPS). As a result, spatial learning and wayfinding of indoor spaces can be far more challenging than performing the same spatial behaviors outside. It is therefore surprising how little research has been done with indoor navigation and how many differences--environmental, technological, and behavioral—are still poorly understood compared to outdoor travel.

Our primary research interest in this topic is to provide solutions to the indoor wayfinding challenge by determining the optimal information requirements for both visual and non-visual interfaces to support spatial knowledge acquisition and navigation of indoor environments. We study this problem using physical spaces, virtual layouts, and mixed environments.

To appreciate the issues involved, consider the last time you walked around a large city. Access to all matter of environmental cues (e.g. a tall building, river, far-off mountain range, etc) likely served as global landmarks to help guide your navigation and structure your mental representation of the space. However, the benefit of these global landmarks to support your spatial behaviors is generally lost when you are navigating inside buildings, as line of sight information is occluded by walls, doors, and limited access to the outside world. For outdoor travel, the consistency of city blocks, naming conventions of streets, and addressing of buildings also provide important spatial cues about your distance traveled, direction of movement, and location in the city.

By contrast, indoor navigation lacks most of these orienting cues, as hallways are not laid out in blocks or given spatially meaningful or uniquely identifiable names. Although rooms usually have numbers, the numbering scheme is often inconsistent and provides far less information than a street address. Without the structure of these naming and numbering conventions, it is easy to get lost during indoor navigation, difficult to give route directions, and hard to determine how your current heading and position relates to other places in the building—ambiguity which complicates spatial updating and wayfinding behavior. Finally, outdoor travel is supported by many external aids such as maps, compasses, and GPS-based navigation systems comprised of large databases of landmarks, streets, and addresses. The latter is of growing interest, as these systems can provide a navigator with real-time route guidance, information about nearby points of interest, the ability to plan detours or reorient if lost, or a means of exploring the city without concern for getting back to a known location. Given the dearth of cues for indoor travel, a similar system for navigating buildings would have many benefits and applications. Unfortunately, indoor GPS tracking is severely limited due to signal attenuation and multipath problems. In addition, access to precise building databases suitable for supporting navigation are rarely available, and even when models do exist, there are no standards for representing/modeling data structures for indoor spaces. As a result, there is currently no viable analogue to GPS navigation systems for providing real-time positioning and building-specific GIS for supporting indoor travel.

While the absence of indoor navigation systems may lead to some frustrations in finding your desired location, wasted time in figuring out where you are and where you want to go, and significantly longer distances traveled as you wander around the building, consulting maps and signs will usually get you to your destination. Now imagine navigating the building--let’s say it’s a large mall--without vision, as would be the case for a person who is blind or visually impaired. In such thought experiments, sighted people usually worry about falling down stairs, bumping into other people, or running into fixed (and usually hard) obstructions such as benches, poles, or walls. However, detecting and avoiding such obstacles is not a significant problem for an independent blind traveler, since long canes and guide dogs are extremely effective mobility tools for performing these spatial operations. Rather, the main challenge for indoor navigation without vision involves lack of the same orientation information which also frustrates visually-based travel. However, these problems are often compounded for blind people, as much of the information to support indoor wayfinding is only accessible visually, e.g. signs, landmarks, floor plans, and you-are-here maps. Thus, as is described further on our Research Interests and Current Projects pages, an important goal of several initiatives in the lab is to determine techniques that can make this information available to people who cannot access it through visual means.

Our take home message is that providing access to context sensitive, navigationally-relevant environmental information will lead to significant improvements for many of the wayfinding challenges experienced during indoor travel, whether performed with or without vision.