Lost in translation (and rotation): Rapid extrinsic calibration for 2D and 3D LIDARs

Abstract

This paper describes a novel method for determining the extrinsic calibration parameters between 2D and 3D LIDAR sensors with respect to a vehicle base frame. To recover the calibration parameters we attempt to optimize the quality of a 3D point cloud produced by the vehicle as it traverses an unknown, unmodified environment. The point cloud quality metric is derived from Rényi Quadratic Entropy and quantifies the compactness of the point distribution using only a single tuning parameter. We also present a fast approximate method to reduce the computational requirements of the entropy evaluation, allowing unsupervised calibration in vast environments with millions of points. The algorithm is analyzed using real world data gathered in many locations, showing robust calibration performance and substantial speed improvements from the approximations.

Unsupervised Intrinsic Calibration of Depth Sensors via SLAM

“…Similarly, Sheehan et al [18] demonstrate unsupervised calibration of a custom Velodyne-like sensor, making use of the known spinning motion of the sensor base to automatically calibrate beam angles, relative laser positions, and timing offsets. In a similar vein, there are several unsupervised methods that demonstrate extrinsic calibration between multiple depth sensors by making use of sensor motion [2,17]. In the realm of RGB-only cameras, Carrera et al [3] show extrinsic calibration between several cameras on a moving platform, and Civera et al [4] compute camera intrinsics using structure from motion.…”

Self-calibration method for a robotic based 3D scanning system

“…By this approach, an accuracy close to the measurement accuracy of the utilized laser sensor (0.05 mm) is achieved. Instead of improving the accuracy of the calibration, the matching of single scans can also be corrected by using the Iterative Closest Point (ICP) algorithm as shown by Borangiu et al [12] or the point cloud entropy as shown by Maddern et al [14].…”

Range Information Characterization of the Hokuyo UST-20LX LIDAR Sensor

“…This can be done in numerous ways, and can use more than one type of sensor. Some common extrinsic calibration procedures use a LiDAR-Camera procedure as outlined in [7][8][9][10], and multiple LiDAR sensors or multiple sensor views as illustrated by [11][12][13][14][15][16], of a fixed target structure for a faster extrinsic calibration prior to operations [17][18][19][20]. These scenarios require the sensors raw data to be correlated into one coherent picture.…”