Robust Multi-Sensor Fusion Positioning Based on GNSS/IMU Using Factor Graph Optimization

Elham Ahmadi, Mahmoud Elsanhoury, Kannan Selvan, Petri Välisuo, and Heidi Kuusniemi

Peer Reviewed

Abstract:	Global Navigation Satellite Systems (GNSS) are essential to modern infrastructure by supporting a wide range of navigational applications and critical operations in various use cases. As reliance on GNSS grows, developing resilient positioning systems that can operate in challenging environments and mitigate the impact of interference remains a key focus for ongoing research. Aiming to enhance positioning accuracy and reliability, in this paper, we propose a loosely coupled (LC) integration of GNSS and inertial measurement units (IMU) using factor graph optimization (FGO). Our approach explores the use of robust loss function within the FGO framework, with a focus on the Huber loss function to mitigate the effects of outliers and noise in GNSS data. We evaluate the performance of our approach using real-world datasets, which provide real-world urban driving scenarios where GNSS signals are prone to degradation. Results demonstrate that our robust FGO (RFGO) method, which incorporates the Huber loss function, outperforms standard FGO (SFGO), which relies on conventional least-squares optimization, by providing more accurate positioning, especially in challenging GNSS environments where outliers and measurement noise degrade performance. Index Terms—GNSS positioning, IMU measurements, robust PNT, factor graph optimization, sensor fusion
Published in:	2025 IEEE/ION Position, Location and Navigation Symposium (PLANS) April 28 - 1, 2025 Salt Lake Marriott Downtown at City Creek Salt Lake City, UT
Pages:	1247 - 1256
Cite this article:	Ahmadi, Elham, Elsanhoury, Mahmoud, Selvan, Kannan, Välisuo, Petri, Kuusniemi, Heidi, "Robust Multi-Sensor Fusion Positioning Based on GNSS/IMU Using Factor Graph Optimization," 2025 IEEE/ION Position, Location and Navigation Symposium (PLANS), Salt Lake City, UT, April 2025, pp. 1247-1256.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In