Abstract

Services based on indoor navigation and localization play crucial roles in people's daily life, and they are even more important for individuals with visual impairments. In this paper, an indoor navigation and localization system CoFINLo with three key techniques: 1) dynamic feature extraction, 2) incremental model update, and 3) fine-grained trajectory generation, is proposed to implement robust indoor localization with friendly interaction design especially for visual impairment groups. Compared with other indoor localization systems, the proposed CoFINLo can overcome the obstacles caused by the diverse sensing devices, complex indoor environments, and fluctuated wireless signal. The proposed system is verified in an indoor office environment and implemented in a large-scale and comprehensive train station. The average 1.39m localization error suggests that CoFINLo can open a new opportunity to implement robust and user-friendly indoor navigation and localization in large and complex indoor environments, which can be a solid foundation for other location-based services.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call