A Location Representation for Generating Descriptive Walking Directions

A Location Representation for Generating Descriptive Walking Directions, Gary Look, Buddhika Kottahachchi, Robert Laddaga, Howard Shrobe

In this study, the authors argue that while many location-aware applications can reason about space at the level of coordinates and containment relationships, these applications have no way to express the semantics that define how a particular space is used. They developed LAIR that generates walking directions that are similar to those a person would give. It is based on an ontology to model both the geographical relationships between spaces as well as the functional purpose of a given space.

Their examination of the directions revealed a useful list of characteristics. As a result of this analysis, their walking guide produce walking directions with the following properties:
- Directions do no use metric distance or cardinal directions.
- Directions are more complex than a sequence of “go-to” and “turn” directives
- Use landmarks to identify places to turn
- Use landmarks to verify travel in the right direction
- Describe the physical spaces a route passes by, or passes through
- Doors are useful landmarks
- Describe hallway intersections