Performance analysis for a novel adaptive algorithm for real-time point cloud ground segmentation

摘要

This paper proposes a fast algorithm for ground segmentation of point clouds. It differentiates ground points from obstacle points from unstructured point clouds. This segmentation consists of performing an orthogonal projection into the horizontal xy plane followed by a top-down area subdivision. This process self-adapts to the point cloud, focusing processing effort on detailed areas. Real-time performance for execution in low cost embedded devices is demonstrated through quantitative and qualitative test results.