Factor-Graph Optimization for Robust Navigation via High-Precision GNSS/IMU Corroboration

Kyle Leland, Clark Taylor, and Frank van Graas

Peer Reviewed

Abstract:	Reliable positioning and time are core components of safe and successful operation in various applications, such as airborne or autonomous navigation. Global Navigation Satellite Systems (GNSS) are a common solution to find a precise position and time estimate. However, GNSS signals may be unreliable in some settings, such as urban or noncooperative environments, due to various error sources including multipath errors, low-visibility of GNSS satellites, or intentional external interference. These error sources may lead to erroneous measurements of GNSS signals that significantly degrade the navigation solution. A factor-graph representation of the GNSS measurements uses measurements over batches of time and enables more resilient state estimation, as demonstrated in [1]. Research into various robust methods has developed variations of factor-graph optimization for identifying and excluding erroneous GNSS measurements in order to find a robust navigation solution, with methods such as robust cost functions [2], dynamic covariance scaling (DCS) [3], and batch covariance estimation [4]. Separately, literature has published methods for the fusion of GNSS measurements with measurements from an inertial measurement unit (IMU) [5]. This paper proposes an approach for resilient navigation through a factor-graph formulation that incorporates high-quality IMU measurements into a robust GNSS factor-graph formulation. The approach effectively enables the corroboration of GNSS measurements based on their consistency with the precise relative-motion information available from a high-quality IMU. The proposed approach is evaluated using data collected from GNSS and IMU sensors on a vehicle traveling in highway, residential, and urban settings. The proposed approach with GNSS and IMU measurements in a robust framework is compared against GNSS-only measurements in a robust factor-graph framework as well as against GNSS and IMU measurements integrated in a basic factor graph without robust modifications. For the robust frameworks, this paper separately evaluates a Huber robust cost function and DCS implementation. In each of the settings in the dataset, the proposed approach demonstrated improved positional accuracy over the alternative evaluated methods.
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:	97 - 104
Cite this article:	Leland, Kyle, Taylor, Clark, van Graas, Frank, "Factor-Graph Optimization for Robust Navigation via High-Precision GNSS/IMU Corroboration," 2025 IEEE/ION Position, Location and Navigation Symposium (PLANS), Salt Lake City, UT, April 2025, pp. 97-104.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In