TEScalib: Targetless Extrinsic Self-Calibration of LiDAR and Stereo Camera for Automated Driving Vehicles with Uncertainty Analysis

Abstract

In this paper, we present TEScalib, a novel extrinsic self-calibration approach of LiDAR and stereo camera using the geometric and photometric information of surrounding environments without any calibration targets for automated driving vehicles. Since LiDAR and stereo camera are widely used for sensor data fusion on automated driving vehicles, their extrinsic calibration is highly important. However, most of the LiDAR and stereo camera calibration approaches are mainly target-based and therefore time consuming. Even the newly developed targetless approaches in last years are either inaccurate or unsuitable for driving platforms.To address those problems, we introduce TEScalib. By applying a 3D mesh reconstruction-based point cloud registration, the geometric information is used to estimate the LiDAR to stereo camera extrinsic parameters accurately and robustly. To calibrate the stereo camera, a photometric error function is builded and the LiDAR depth is involved to transform key points from one camera to another. During driving, these two parts are processed iteratively. Besides that, we also propose an uncertainty analysis for reflecting the reliability of the estimated extrinsic parameters. Our TEScalib approach evaluated on the KITTI dataset achieves very promising results.

A Novel and Simplified Extrinsic Calibration of 2D Laser Rangefinder and Depth Camera

“…References [25][26][27][28][29][30][31] are the calibration without target, which are further divided into feature-based extrinsic calibration and motion-based extrinsic calibration. Levison et al [25] proposed a self-calibration method based on edge feature matching.…”

“…However, such methods rely on image feature points and radar point cloud data that are often difficult to obtain in natural scenes and have harsh usage conditions. In order to estimate the LiDAR to stereo camera extrinsic parameters for driving platforms, applying 3D mesh reconstruction-based point cloud registration, a photometric error function was built [31]. In addition to directly obtaining the extrinsic parameter of calibration, CFNet proposed by Wang [32] was utilized to predict the calibration flow based on convolutional neural networks.…”