Magnetic Maps of Indoor Environments for Precise Localization of Legged and Non-legged Locomotion

Abstract

The magnetic field in indoor environments is rich in features and exceptionally easy to sense. In conjunction with any form of odometry, such as wheel- or IMUbased, a map of this field can be used to precisely localize a human or robot in indoor environments. We show how the use of this field yields significant improvements in terms of accuracy and computational complexity for both legged and wheel-based locomotion. We suggest various likelihood functions for sequential Monte Carlo algorithms for localization and evaluate their performance based on magnetic maps of several environments. Specifically, we investigate the influence of intensity and vector-based representations as well as map resolution on localization accuracy, robustness and complexity. Compared to other localization approaches, such as camera or LIDAR-based, far less privacy concerns exist for the magnetic field. Furthermore, the required sensors are less costly, have a lower raw data rate, yet appear comparably informative for localization purposes. This combination of technical and privacy-related advantages renders localization of humans and robots based on the magnetic field a very viable solution to indoor navigation.