PERSONAL NAVIGATION: EXTENDING MOBILE MAPPING TECHNOLOGIES INTO INDOOR ENVIRONMENTS
DOI:
https://doi.org/10.5380/bcg.v15i5.17000Palavras-chave:
Multi-sensor Navigation, Dead Reckoning, AI Methods, Human Locomotion Modeling, Image-based Navigation.Resumo
This paper discusses some unconventional methods for indoor-outdoor navigation, based on the integration of self-contained sensors, including GPS, IMU, digital barometer, magnetometer compass, and a human locomotion model. The human
locomotion model is used as navigation sensor and it is handled by Artificial Intelligence (AI) techniques that form an adaptive knowledge-based system (KBS), which is trained during the GPS signal reception, and is used to support navigation under GPS-denied conditions. A complementary technique used in our solution, which facilitates indoor navigation, is the image-based method (Flash LADAR). In this paper, the system design and an example performance analysis in the mixed indoor-outdoor environment are presented.