Emphasis of LiDAR data fusion using iterative closest point and ICP registration

Light detection and ranging (LiDAR) is an optical remote-detecting technique that utilizes laser light to densely sample the earth's surface, delivering exceptionally exact x, y, z estimations. Lieder, essentially utilized in airborne laser mapping applications, is developing as a financially savvy option in contrast to conventional looking over strategies, for example, Photogrammetry. Spatially sorted post prepared LiDAR information is known as point cloud data (PCD) and LiDAR produces these point cloud datasets in mass. The underlying point mists are vast accumulations of 3D height focuses, which incorporate x, y, z and intensity. Data can be collected from multiple LiDARs simultaneously, which are placed adjacent to each other like one on each of the headlights in front of the car. Each LiDAR produces its own PCD. A method is required to combine the PCD from participating LiDARs for further analysis. In this paper, a novel approach for combining the data from multiple LiDAR is proposed, which involves obtaining inliers and outliers. The efficient algorithms such as conversion algorithm, Iterative Closest Point and ICP registration are used in the process. The conversion algorithm is applied to point cloud data to render the mesh representation of the actual image from PCD. Outliers are separated using distance algorithms such as Euclidean and are discarded. Inlier data are treated with Iterative Closet Algorithm (ICP) to generate a matrix of points. Finally, ICP registration has been applied to consecutive data frames from adjacent LiDARs and combined PCD is retrieved resulting in the fusion of PCDs. The LiDAR data fusion has various applications in the field of autonomous driving.