Use of PCL library to obtain the Transformation Matrix. State of the Art

Sergio Cebollada López

Resumen


Map building and localization are two of the most important topics in autonomous mobile robots. In this sense, researchers have proposed many ways to solve these problems. One of the methods used for this purpose is through the point clouds obtained by 3D scanners. The Registration is used to know the movement of the robot into the map.  Given two point clouds, this tool returns the Transformation Matrix. This work presents a state of the art about the tools implemented in the Point Cloud Library. Two big methods are used to make the Registration, the first one is based in alignment between keypoints from the point clouds, the second one consists in iterative registration algorithms. Both methods are presented in this work as well as several possible tools implemented for this purpose until now. Finally, a comparison between both methods is carried out. This work has been performed to gain access to TECNIT doctoral program at Universidad Miguel Hernndez de Elche and supported the Spanish government through the project DPI2013-41557-P.


Texto completo:

PDF



ISSN: 2530-7320


Creative Commons License
Esta obra está bajo una licencia de Creative Commons Reconocimiento-
NoComercial-CompartirIgual 4.0 Internacional
.