Android GNSS/INS Using Complementary Filter

Dong-Kyeong Lee, Evan Gattis, Dennis Akos, Jeonghyeon Yun, Byungwoon Park

Peer Reviewed

Abstract: Since the availability of raw GNSS measurements in Android devices in 2016, the smartphone users have acquired the ability to not only process the GNSS position, velocity, and time solutions from the chipsets, but also utilize the raw measurements to carry out their own GNSS post-processing. Consequently, rather than relying on the black-box algorithms inside the chipsets, it has become possible to compute the solutions in the measurement domain. In addition, due to the nature of smartphones, there are additional navigation sensors available including inertial navigation sensors (INS). However, most of the publicly available literature and smartphone positioning challenges have had limited success at combining the GNSS measurements with non-GNSS sensors. This paper outlines in detail how the inertial sensors can be fused with the GNSS measurements, and coupled with differential GNSS (DGNSS), single differenced navigation algorithms, and tightly coupled extended Kalman Filter (EKF), can achieve 1.7 m RMS position accuracy even in urban canyon environments. The attitude computation and bias estimation for the INS are carried out using the complementary filter instead of inside the EKF. This means that the attitudes are less affected by unmodeled GNSS measurement anomalies, and the EKF state estimation can be reduced from the typical 17 or more states to 6 states. The complementary filter is a popular approach for drone navigation, and it is especially helpful for smartphone devices that are prone to GNSS anomalies in urban canyon environments. The paper will outline in detail how the INS measurements can be combined with raw GNSS measurements in any smartphone using the complementary filter, and how challenges introduced with a fewer state EKF can be mitigated. Finally, the results from an urban canyon drive test carried out in downtown Denver, Colorado will be presented.
Published in: Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023)
September 11 - 15, 2023
Hyatt Regency Denver
Denver, Colorado
Pages: 2644 - 2652
Cite this article: Lee, Dong-Kyeong, Gattis, Evan, Akos, Dennis, Yun, Jeonghyeon, Park, Byungwoon, "Android GNSS/INS Using Complementary Filter," Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023), Denver, Colorado, September 2023, pp. 2644-2652.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In