Lie Algebra Based Augmented State EKF Design For Information Fusion in Odometry

Date

2025-1-10

Author

Yalçın, Haktan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

40
views

85
downloads

Cite This

Rigid transformations or rotation matrices, which inherently do not belong to any vector space, are frequently encountered in state estimation. However, many state estimation frameworks are designed to operate on vector spaces. This thesis explores the application of Lie Algebra to effectively handle rigid transformations within state estimation. Specifically, it examines methods for expressing the uncertainty associated with rigid transformations and for differentiating nonlinear functions with respect to rotation matrices. As a key contribution, an augmented state extended Kalman filter is developed to integrate incremental pose information derived from a stereo camera setup with inertial measurements. The proposed framework is evaluated using the publicly available KITTI dataset. Experimental results demonstrate the effectiveness of the fusion in obtaining more accurate trajectory estimation.

Subject Keywords

Lie Algebra, Extended Kalman Filter, State Estimation, Visual Inertial Odometry, Dense Odometry

URI

https://hdl.handle.net/11511/113439

Collections

Graduate School of Natural and Applied Sciences, Thesis