Version 1
: Received: 30 September 2024 / Approved: 1 October 2024 / Online: 1 October 2024 (15:32:36 CEST)
How to cite:
Tian, K.; Cheok, K.; Radovnikovich, M.; Kobayashi, K.; Cai, C. Singular Value Decomposition (SVD) Method for LiDAR and Camera Sensor Fusion and Self-Calibration Algorithm. Preprints2024, 2024100081. https://doi.org/10.20944/preprints202410.0081.v1
Tian, K.; Cheok, K.; Radovnikovich, M.; Kobayashi, K.; Cai, C. Singular Value Decomposition (SVD) Method for LiDAR and Camera Sensor Fusion and Self-Calibration Algorithm. Preprints 2024, 2024100081. https://doi.org/10.20944/preprints202410.0081.v1
Tian, K.; Cheok, K.; Radovnikovich, M.; Kobayashi, K.; Cai, C. Singular Value Decomposition (SVD) Method for LiDAR and Camera Sensor Fusion and Self-Calibration Algorithm. Preprints2024, 2024100081. https://doi.org/10.20944/preprints202410.0081.v1
APA Style
Tian, K., Cheok, K., Radovnikovich, M., Kobayashi, K., & Cai, C. (2024). Singular Value Decomposition (SVD) Method for LiDAR and Camera Sensor Fusion and Self-Calibration Algorithm. Preprints. https://doi.org/10.20944/preprints202410.0081.v1
Chicago/Turabian Style
Tian, K., Kazuyuki Kobayashi and Changqing Cai. 2024 "Singular Value Decomposition (SVD) Method for LiDAR and Camera Sensor Fusion and Self-Calibration Algorithm" Preprints. https://doi.org/10.20944/preprints202410.0081.v1
Abstract
Lidar and camera sensors are widely utilized in autonomous vehicles (AVs) and robotics due to their complementary sensing capabilities—lidar provides accurate depth information, while cameras capture rich visual data. However, effective fusion of these sensors requires precise calibration, which is often challenging due to the differences in sensing modalities. This paper presents a self-calibration algorithm based on Singular Value Decomposition (SVD) to address this problem. Singular Value Decomposition is a powerful matrix factorization technique that helps identify key calibration parameters by decomposing matrices into singular vectors and values, representing essential calibration data. Our method not only automates the calibration process but also provides a reliable mechanism for detecting calibration errors, ensuring long-term accuracy and robustness in sensor fusion. The algorithm simplifies deployment in real-world applications by eliminating the need for manual calibration or external targets. Experimental results show that our approach improves sensor alignment and fusion accuracy, making it particularly useful for AV and robotic systems where maintaining precise calibration is critical.
Keywords
LiDAR and Camera data sensor fusion; Singular Value Decomposition; Self-calibration; error detection
Subject
Engineering, Electrical and Electronic Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
