Precise Positioning of Smartphones Using a Robust Adaptive Kalman Filter

Anurag Raghuvanshi, Sudha Vana, Sunil Bisnath

Peer Reviewed

Abstract:	Android smartphones equipped with multi-constellation and dual-frequency GNSS chipsets are seen as a significant portion of the future for low-cost GNSS-based applications. These smartphones have already demonstrated good performance in applications such as pedestrian navigation, fitness tracking, and location-based services. Smartphones equipped with GNSS chips are considered an essential component in future high-accuracy applications. The low-cost GNSS antenna and receiver in a smartphone make achieving high accuracy a challenging problem to solve. However, recent advancements in accuracy are evident from the literature on the topic, and the community is striving for decimeter-level accuracy. Google is spearheading the effort with its Google Decimeter Challenge, and participants in the challenge have shown significant improvements in accuracy using various techniques like Precise Point Positioning (PPP), Real-Time Kinematic (RTK), and machine learning. These techniques leverage advanced algorithms and data processing methods to enhance the positioning accuracy of GNSS-enabled smartphones. The current effort is a step towards achieving decimeter-level accuracy for smartphones. A robust adaptive Kalman filter (RAKF) is used in this research to improve the tightly-coupled (TC) GNSS and IMU solution for smartphones. The RAKF technique has not been explored so far in the literature with respect to smartphone measurements. Measurement and dynamic model adaptiveness are adopted in this research to balance the contributions of measurements and predicted states. The standardized measurement residuals (innovation) and ratio of variance is used as a learning statistic to apply the adaptiveness to the measurements and the dynamic model of the Kalman filter, respectively. The RAKF technique improves the overall root mean square error by 6-28%, a positive step towards better smartphone accuracy.
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:	2749 - 2764
Cite this article:	Raghuvanshi, Anurag, Vana, Sudha, Bisnath, Sunil, "Precise Positioning of Smartphones Using a Robust Adaptive Kalman 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. 2749-2764. https://doi.org/10.33012/2023.19296
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In