Path markup language for indoor navigation

Abstract

Indoor navigation is critical in many tasks such as firefighting, emergency medical response, and SWAT response, where GPS signals are not available. Prevailing approaches such as beacons, radio signal triangulation, SLAM, and IMU methods are either expensive or impractical in extreme conditions, e.g. poor visibility and sensory drifting. In this study, we develop a path markup language for pre-planning routes and interacting with the user on a mobile device for real-time indoor navigation. The interactive map is annotated with walkable paths and landmarks that can be used for inertial motion sensor-based navigation. The wall-following and landmark-checking algorithms help to cancel drifting errors along the way. Our preliminary experiments show that the approach is affordable and efficient to generate annotated building floor path maps and it is feasible to use the map for indoor navigation in real-time on a mobile device with motion sensors. The method can be applied to intelligent helmets and mobile phones, including potential applications of first responders, tour guide for buildings, and assistance for visually impaired users.