| IP개요 |
LiDAR point cloud clustering is a crucial part of object detection and recognition. However, clustering enormous point cloud of LiDAR assigns a large computation loads to autonomous robot with low-power core. In this paper, we propose a low-power point cloud clustering system with a density-based spatial clustering of applications with noise (DBSCAN) clustering algorithm based on Cortex-M0. For accelerating the clustering and alleviating the computational loads, the accelerator is embedded on system. To verify the feasibility of the accelerator, we implemented the point cloud clustering accelerator on a field programmable gate array (FPGA) with 2989 frames of Pixset dataset. The clustering speed is enhanced 7.5x compared HW accelerator design to SW design. |
