 |
| Titel |
Automated science target selection for future Mars rovers: A machine vision approach for the future ESA ExoMars 2018 rover mission |
| VerfasserIn |
Yu Tao, Jan-Peter Muller |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250079780
|
|
|
|
|
|
| Zusammenfassung |
| The ESA ExoMars 2018 rover is planned to perform autonomous science target selection
(ASTS) using the approaches described in [1]. However, the approaches shown to date have
focused on coarse features rather than the identification of specific geomorphological units.
These higher-level “geoobjects” can later be employed to perform intelligent reasoning or
machine learning. In this work, we show the next stage in the ASTS through examples
displaying the identification of bedding planes (not just linear features in rock-face images)
and the identification and discrimination of rocks in a rock-strewn landscape (not just rocks).
We initially detect the layers and rocks in 2D processing via morphological gradient
detection [1] and graph cuts based segmentation [2] respectively. To take this further
requires the retrieval of 3D point clouds and the combined processing of point clouds
and images for reasoning about the scene. An example is the differentiation of
rocks in rover images. This will depend on knowledge of range and range-order of
features.
We show demonstrations of these “geo-objects” using MER and MSL (released through
the PDS) as well as data collected within the EU-PRoViScout project (http://proviscout.eu).
An initial assessment will be performed of the automated “geo-objects” using the
OpenSource StereoViewer developed within the EU-PRoViSG project (http://provisg.eu)
which is released in sourceforge. In future, additional 3D measurement tools will be
developed within the EU-FP7 PRoViDE2 project, which started on 1.1.13.
References: [1] M. Woods, A. Shaw, D. Barnes, D. Price, D. Long, D. Pullan, (2009)
“Autonomous Science for an ExoMars Rover–Like Mission”, Journal of Field Robotics
Special Issue: Special Issue on Space Robotics, Part II, Volume 26, Issue 4, pages 358-390.
[2] J. Shi, J. Malik, (2000) “Normalized Cuts and Image Segmentation”, IEEE Transactions
on Pattern Analysis and Machine Intelligence, Volume 22. [3] D. Shin, and J.-P. Muller
(2009), Stereo workstation for Mars rover image analysis, in EPSC (Europlanets), Potsdam,
Germany, EPSC2009-390 |
|
|
|
|
|
|