Reliable real-time extrinsic parameters of 3D Light Detection and Ranging (LiDAR) and cameras are vital components of multimodal perception systems. However, extrinsic transformation may drift gradually during operation, resulting in decreased accuracy of the perception system. To solve this problem, this chapter proposes a line-based method that enables automatic online extrinsic calibration of LiDAR and cameras in real-world scenes. Moreover, adaptive optimization is utilized to provide accurate extrinsic parameters. Experiments verify that the proposed approach automatically corrects miscalibration errors and achieves an accuracy of 0.2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$^{\circ }$$\end{document} . This work can provide a basis for perception systems and further improve the performance of other algorithms that utilize these sensors.
Multi-Sensor Calibration
Multi-sensor Fusion for Autonomous Driving ; Chapter : 3 ; 37-49
2023-05-11
13 pages
Article/Chapter (Book)
Electronic Resource
English