Low data loss point cloud to multi-line conversion and union

článek ve sborníku ve WoS nebo Scopus

en

Union of point clouds converted into lines is basically impossible nowadays. In big area mapping task, there is a problem how to represent scanned data. A point cloud is basic output from almost every mapping and environment scanning system, but as the area grows, there comes need for huge data storage and also the computation cost starts to grow rapidly. On the other side, curve representation is simple and easy to process, but all other data from point cloud get lost and further actualization of curve is basically impossible. In this article we present algorithms of robust procedure which can transfer point-cloud to curve and data. The final goal is to achieve the same result for two curves which are merged, as if two point-clouds were merged and transferred to curve.

Union of point clouds converted into lines is basically impossible nowadays. In big area mapping task, there is a problem how to represent scanned data. A point cloud is basic output from almost every mapping and environment scanning system, but as the area grows, there comes need for huge data storage and also the computation cost starts to grow rapidly. On the other side, curve representation is simple and easy to process, but all other data from point cloud get lost and further actualization of curve is basically impossible. In this article we present algorithms of robust procedure which can transfer point-cloud to curve and data. The final goal is to achieve the same result for two curves which are merged, as if two point-clouds were merged and transferred to curve.

point cloud, line approximation, mapping, SLAM, multi-agent system, multi-robot system

02.12.2020

978-1-7281-5600-2

Proceedings of the 2020 19th International Conference on Mechatronics - Mechatronika (ME)

1–4

4