The article presents a method for extracting planar regions in uneven terrains from unordered point cloud measurements, which is crucial for robotic locomotion. The proposed multi-resolution planar region extraction strategy aims to balance accuracy and computational efficiency. The approach starts with a pointwise classification preprocessing module that categorizes all points based on their local geometric properties. The points are then arranged using an octree, followed by an in-depth analysis of nodes to finish the multi-resolution plane segmentation. The method’s efficiency and robustness are verified through synthetic and real-world experiments, demonstrating its ability to generalize effectively across various uneven terrains while maintaining real-time performance.
Publication date: 22 Nov 2023
Project Page: N/A
Paper: https://arxiv.org/pdf/2311.12562
