Over the past few years, several technological advances have been made to enable locating people in indoor settings, where way finding is something we do on a daily basis. In a similar way as it happened with GPS and today's popular outdoor navigation systems, indoor navigation is set to become one of the first, truly ubiquitous services that will make our living and working environments intelligent. Two critical characteristics of human way finding are destination choice and path selection. This work focuses on the latter, which traditionally has been assumed to be the result of minimizing procedures such as selecting the shortest path, the quickest or the least costly path. However, this path approximations are not necessarily the most natural paths. Taking advantage of context-aware information sources, this paper presents an easy to deploy context-aware indoor navigation system, together with an efficient spatial representation, and novel approach for path adaptation to help people find their destination according to their preferences and contextual information. We tested our system in one building with several users to estimate first an assessment of preference values, and later to compare how the paths suggested by our system correspond to those people would actually follow. The positive results of this evaluation confirm the suitability of our models and algorithms.
展开▼
