Statistics – Computation
Scientific paper
Dec 2002
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002agufm.p22a0394l&link_type=abstract
American Geophysical Union, Fall Meeting 2002, abstract #P22A-0394
Statistics
Computation
6225 Mars
Scientific paper
High precision topographic information is critical to many landing site geological and engineering applications. Precise navigation and localization of the Mars rover is important both for its own safety as well as for its ability to accomplish engineering and scientific objectives as it traverses the Martian surface. Thus high precision landing site mapping and rover localization is very desirable for the support of future long-range rover missions such as the 600-meter to 1,000-meter traverse planned for the 2003 MER mission. We have developed algorithms and software for the integrated bundle adjustment of ground images. An incremental bundle adjustment model has also been developed that adjusts descent and rover images in a progressive process that results in increased computational efficiency. An innovative approach has been investigated for automatic feature extraction and tie-point selection based on interesting point filtering and image matching techniques. Two field tests were conducted (April 1999 and May 2000) at Silver Lake, CA. Various rover localization experiments were carried out. Using descent and rover images and either an integrated or incremental adjustment, rover localization accuracy of one percent was achieved of about 1m for a traverse length of 1km from the landing center. Experiment results also showed that if no descent images are available (as will be the case in the 2003 MER mission), it is still feasible to localize a rover using only rover images. In addition to using simulated descent and rover images, we tested our methods and software with actual Mars data - IMP lander (Imager for Mars Pathfinder) and rover images form 1997 Mars pathfinder mission. With the bundle adjustment, the image errors were reduced from several - tens of pixels to a sub-pixel level. This indicates that the bundle adjustment has improved the exterior orientation (EO) parameters significantly. Seamless DEM and orthoimage can then be generated using the improved EO parameters. We are currently processing data from the Athena Science Team's FIDO test conducted in August 2002. In this test, MOC, THEMIS and MOLA data were simulated for the FIDO site and actual Navcam and Pancam images were taken. We will present the latest results at the meeting.
Arvidson Ray E.
Di Ke
Li Renwei
Matthies Larry H.
Olson Clark F.
No associations
LandOfFree
Ground Image Based High Precision Mars Rover Localization and Landing Site Mapping does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Ground Image Based High Precision Mars Rover Localization and Landing Site Mapping, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Ground Image Based High Precision Mars Rover Localization and Landing Site Mapping will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1892000