OSIRIS REx NASA Mission
A SPC rendered shape model of Bennu
The accuracy of the global digital terrain model (GDTM) of Bennu obtained by the OSIRIS-REx mission to date has achieved global requirements earlier than anticipated. Additionally, a comparison between real images and synthetic images derived from the GDTM using normalized cross correlation show an exceptionally good match.
101955 Bennu Shape Model PDS SBN OSIRIS-REx SPICE Data Archive Publications:
Image Cross Correlation as a Measurement of Shape Model Quality
Reference Citation: Palmer, E.~E., Weirich, J.~R., Campbell, T., Barnouin, O.~S., Daly, M.~G., Lauretta, D.~S.\ 2018.\ Image Cross Correlation as a Measurement of Shape Model Quality.\ Lunar and Planetary Science Conference 1106.
Stereophotoclinometry Models in Support of the OSIRIS-REx Mission .
Reference Citation: Palmer, E.~E., and 11 colleagues 2019.\ Stereophotoclinometry Models in Support of the OSIRIS-REx Mission.\ Lunar and Planetary Science Conference 2588.
Evaluation of the Bennu Global Digital Terrain Model for OSIRIS-REx
Reference Citation: Palmer, E., and 7 colleagues 2019.\ Evaluation of the Bennu Global Digital Terrain Model for OSIRIS-REx.\ EPSC-DPS Joint Meeting 2019 2019, EPSC-DPS2019-387.
Evaluation of the Bennu Global Digital Terrain Model for OSIRIS-REx EPSC AbstractsVol. 13, EPSC-DPS2019-387-1, 2019 EPSC-DPS Joint Meeting 2019
Stereophotoclinometry Models in Support of the OSIRIS-REx Mission 50th Lunar and Planetary Science Conference 2019 (LPI Contrib. No. 2132)
Image Cross Correlation as a Measurement of Shape Model Quality 49th Lunar and Planetary Science Conference 19-23 March, 2018, held at The Woodlands, Texas LPI Contribution No. 2083, id.1106
Evaluation of Topographic Error and Quality with Stereophotoclinometry 49th Lunar and Planetary Science Conference 2018 (LPI Contrib. No. 2083)
Error Analysis of Stereophotoclinometry in Support of the OSIRIS-REx Mission American Astronomical Society, DPS meeting #47, id.213.17
A progression of the SPC process throughout the OSRIS-REx mission
2013 Nolan Model
The first shape model was done by Dr. Nolan and was based on radar imaging of Bennu (Nolan, 2013). This model was the basic reference for the mission planning of OSIRIS-REx. One can see that the basic shape of Bennu was well characterized. The comparison image was from early November 2018. 12 November Early Limb-based model
Altimetry Working Group used early Approach Phase rotation images, every 10°, to determine the limbs and construct a model with a 3m ground sample distance. It improved the accuracy of the shape model, and Navigation Team had exceptional success with its center-finding routines. 22 November Early SPC model
While SPC can be used earlier, we used the limb-based model until the imaging data was exceptional. This reduces artifacts and improves accuracy for following models. OSIRIS-REx approached the asteroid at a low phase, so there are limited shadows from the topography. 2nd December – End of approach
At the end of the Approach Phase, a 1.5m model was generated from the full suite of approach images. Limb, albedo and topography agreement was very good; however, the poles lacked imaging so there is only limb-based data there. Final imagery pixel size was 32cm near the equator. End of PreSurvey Model
At the end of preliminary Survey, we generated a model with a 75cm ground sample distance. Both the topography and albedo representation were excellent and exceeded our expectations. Preliminary Survey data was 50-75 cm in pixel size.
Here is a regional model at the end of Preliminary Survey, with 50 to 75 cm resolution. This is of DL15 aka Nightingale aka the Primary TAG site. Images below are the progression as we processed more images.
14 cm model resolution of the Primary TAG site.
5 to 6 cm resolution of the Primary TAG site.
State of the model after Natural Feature Tracking (NFT) landmarks have been generated. Some regions have high resolution (up to 1.7 cm) while others are the same as in the above image.
Animate GIF showing the comparison between spacecraft image and model for Primary TAG site. 6 second loop.