Anzeige
Erschienen in:
2016 | OriginalPaper | Buchkapitel
Rendering of 3D Maps with Additional Information for Operator of a Coal Mine Mobile Robot
verfasst von : Tomáš Kot, Petr Novák, Jan Babjak, Petr Olivka
Erschienen in: Modelling and Simulation for Autonomous Systems
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
The paper focuses on visualization of point clouds made by a 3D scanner mounted on a mobile robot Telerescuer designed for reconnaissance of coal mines affected by a disaster. Briefly are described some algorithms used for point cloud pre-processing – voxelization for data reduction, outliers removing for filtering of erroneous data and smoothing for additional filtering of noise data. These algorithms are implemented in C++ using the Point Cloud Library.
The next parts focus on the rendering engine created for this application, with more detailed information about drawing individual points with specific size and using the point colours to support better representation of shapes in the map by shading/lighting and additional colouring based on orientation of normal vectors. Mentioned are also some crucial optimizations of rendering and processing performance build on a simple custom system similar to Octree.
The final part presents some methods of adding additional information to the map, including sensor readings (temperature, gas concentration, wind speed etc.) and distance measurements (exact numeric measuring, rough dimension estimation by colour coding, corridor cross-section etc.). Integration of these data and the advanced rendering techniques not typically used for point cloud visualization are the innovative approaches described in this paper.