Factor Graph Dimensionality Reduction Using Lateral Motion Constraints for Aided Dead Reckoning Navigation

Adam Rutkowski, Yetong Zhang, Frank Dellaert

Abstract: Abstract—For navigation problems involving dead reckoning of odometry measurements aided with additional sensors, we introduce a method that treats the lateral components of the odometry measurements as constraints, thereby reducing the dimensionality of the state representation. The constrained lateral motion approach is best suited for factor graph representations of ground vehicle and fixed-wing aerial vehicle navigation, whereby the tangential component of motion is typically much greater than the lateral component. We conduct experiments in both 2D and 3D cooperative navigation scenarios aided by inter-vehicle range measurements, and show that we achieve faster convergence with more efficient optimization with our new parameterization. Index Terms—motion constraints, state estimation, cooperative navigation, constrained optimization
Published in: 2023 IEEE/ION Position, Location and Navigation Symposium (PLANS)
April 24 - 27, 2023
Hyatt Regency Hotel
Monterey, CA
Pages: 755 - 763
Cite this article: Rutkowski, Adam, Zhang, Yetong, Dellaert, Frank, "Factor Graph Dimensionality Reduction Using Lateral Motion Constraints for Aided Dead Reckoning Navigation," 2023 IEEE/ION Position, Location and Navigation Symposium (PLANS), Monterey, CA, April 2023, pp. 755-763. https://doi.org/10.1109/PLANS53410.2023.10140049
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In