Navigation through the Processing of Android Data with a High-Order Kalman Filter

Christian J. Campos-Vega, Tanner M. Watts, Scott M. Martin, Howard Chen, and David M. Bevly

Abstract: This paper presents a fourth order Extended Kalman Filter (EKF) that processes Android smartphone data for positioning and navigation. The EKF utilizes GNSS, IMU, and magnetometer sensor outputs for correction to its states. Unlike traditional GPS/INS error state recursions, the EKF processes the IMU outputs through the measurement observation matrix. The highorder EKF includes Fault Detection and Exclusion (FDE) to remove erroneous measurements. The navigation algorithm was evaluated with Pixel 4 and Pixel 4XL training data provided by the Google Smartphone Decimeter Challenge. The experimental results indicate the smartphones’ GNSS chip receivers provide positions with accuracies of roughly 1.5 meters at a rate of 1 Hz. The high-order EKF provides positions with accuracies of roughly 3 meters at a rate of 100 Hz. Furthermore, the highorder EKF can accurately navigate under motion with the IMU and magnetometer sensors for 5 to 10 seconds without GNSS. In cases when the GNSS has position faults, the EKF successfully rejects the outliers with FDE and continues to navigate. Future work includes augmenting the EKF with an adaptive process noise tuning algorithm, including sensor bias states for the IMU and magnetometer sensors, and comparing the high-order EKF to a traditional GPS/INS error state filter.
Published in: Proceedings of the 34th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2021)
September 20 - 24, 2021
Union Station Hotel
St. Louis, Missouri
Pages: 2957 - 2973
Cite this article: Campos-Vega, Christian J., Watts, Tanner M., Martin, Scott M., Chen, Howard, Bevly, David M., "Navigation through the Processing of Android Data with a High-Order Kalman Filter," Proceedings of the 34th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2021), St. Louis, Missouri, September 2021, pp. 2957-2973. https://doi.org/10.33012/2021.18042
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In