DATA_SET_DESCRIPTION |
Data Set Overview
=================
The HAYABUSA spacecraft included a LIght Detection and Ranging (LIDAR)
altimeter.The primary objective of LIDAR was to establish the range between
the HAYABUSA spacecraft and the asteroid Itokawa for navigation purposes
during the surveying and collection phases of the mission. It provided
excellent estimates of the location of the spacecraft relative to the
asteroid.
The secondary scientific objective of the LIDAR included determining the
mass of the asteroid, and measuring its global surface elevation and
roughness.
The Calibrated Data Record (CDR) contains LIDAR science and telemetry data
that have been converted to engineering and physical units. The Experiment
Data Record (EDR) is the source for the science data, while the House
Keeping Experiment Data Record (HKEDR) and telemetry provide the data needed
to determine the position of the HAYABUSA spacecraft relative to the
asteroid. Resulting orbit, geometric, and calibration data have been
incorporated, to determine the location of the LIDAR boresight on the
surface of the asteroid provided in the CDR. With an appropriate shape model
and density estimates for Itokawa, these can be used to obtain topographic
profiles of the surface, e.g. [CHENGETAL2002] of Itokawa.
The confirmed time resolution of the HAYABUSA lidar equals aproximately
1.67ns which is equivalent to 0.5m in range. This values was derived from
the data obtained over the very flat and smooth Muses-C Regio. See
[BARNOUIN-JHAETAL2008]. For further information about the Hayabusa LIDAR and
its data set may be found in [MUKAIETAL2006] and [BARNOUIN-JHAETAL2008].
Data
====
All the HAYABUSA LIDAR data records are ascii tables. Each EDR contains
three columns of data. The first column is the the Mission Elapsed Time in
units of spacecraft ticks when the LIDAR range was measured. One spacecraft
tick equals 1/32 of a second. The second column is equal to this same time
but translated into units of Universal Coordinated Time UTC. The third
column equals the range measured by the LIDAR at the time indicated. The
LIDAR collected data at 1 Hz (1 return per second) for the entire duration
of the encounter with Itokawa (~3 months).
The EDR is composed of three files: EDR20050911_20050929.TAB which
corresponds to data acquired during the GATE Position Phase when the
HAYABUSA spacecraft was at ~20km distance from the surface of the asteroid;
EDR20050930_20051028.TAB which corresponds to data acquired during the HOME
Position Phase when the HAYABUSA spacecraft was at 3-7km distance from the
surface of the asteroid; EDR20051029_20051125.TAB which corresponds to the
TOUCH DOWN phase when the HAYABUSA spacecraft attempted to sample the
surface of Itokawa.
The housekeeping (HKEDR) file has not been included in the archive because
the mission has not yet given permission for it to be archived in PDS. When
permission is given to archive it, it will be added. The first three
columns of this HKEDR are identical to those in the EDR file. The next two
columns provide the illuminated centroid of Itokawa within the reference
frame of the wide angle camera, ONC-W1. This camera is part of the AMICA
instrument package and more information is given in the description for that
instrument. In brief, it is a 512 by 512 pixel imager with a field of view
of 60 degrees.
The software aboard the HAYABUSA spacecraft imager found the centroid of
the largest illuminated object in the field of view of this ONC-W1, with DN
values greater than a threshold of 5DN. This happened at 2 min intervals,
and was accompanied by a LIDAR range and time listed in the first three
columns of the HKEDR file already discussed. The first of the two pixel
values gives the x or sample location of the centroid. The second one gives
the y or line location of the centroid. The one HKEDR file includes data for
the entire mission. A pixel value for x and y of zero implies the asteroid
filled the entire field of view of the ONC-W1.
The CDR data files are of three kinds. The first is an unfiltered (UF)
version of the data after the processing described in the next section was
undertaken. The second is composed of the same set but was filtered (F) to
remove any estimated surface point which was located more that 10m from the
predicted intersection of the vector defining the pointing of the LIDAR
boresight and the shape model of the asteroid. The third is an optimized
(OPT) version that is built on the filtered data. The filenames include F,
UF and OPT respectively. The CDR files include information on both the
HAYABUSA LIDAR and the boresight of the Near Infra-Red Spectrometer aboard
HAYABUSA. The NIRS data is provided as a courtesy to others HAYABUSA related
efforts because it is exactly aligned with the LIDAR.
All three filtered, unfiltered and optimized CDR files list the following
data in their column order:
1. The spacecraft mission elapsed time (MET)
2. The equivalent spacecraft time in UTC
3. The X,Y and Z estimate of the spacecraft location derived during the
LIDAR data processing (see processing section below)
4. The estimated X,Y,Z position of the LIDAR and Near Infra-Red
Spectrometer (NIRS; Co-aligned with the LIDAR) footprint using the LIDAR
processed spacecraft position, the LIDAR boresight vector, and the measured
range.
5. The predicted X,Y,Z position of the LIDAR footprint at the
intersection of the LIDAR boresight vector with the a high resolution
asteroid shape model [GASKELLETAL2008B] using the LIDAR processed spacecraft
position.
6. Incidence, emission and phase angle of the center of the LIDAR/NIRS
field of view (FOV) using the interection of the LIDAR boresight vector with
a high resolution asteroid shape model [GASKELLETAL2008B], using the LIDAR
processed spacecraft position.
7. Size in m of the LIDAR and NIRS FOV obtained from the LIDAR FOV and
measured LIDAR range to surface of the asteroid.
8. Longitude and latitude of the center of the LIDAR/NIRS (FOV) using
the measured LIDAR range, and the spacecraft positions given by the LIDAR
processing.
9. Predicted longitude and latitude of the LIDAR/NIRS (FOV) at the
intersection of the LIDAR boresight vector with the asteroid shape model
[GASKELLETAL2008B], using the spacecraft positions given by the LIDAR
processing.
10. Mean incidence, emission and phase angle from nine locations across
the FOV of NIRS that were derived from nine vectors across the NIRS FOV
(split into a 3x3 grid) whose intersection with the shape model
[GASKELLETAL2008B] were determined using the spacecraft positions given by
the LIDAR processing.
11. Mean longitude and latitude of the NIRS FOV for the up to nine
vectors within the NIRS FOV that intersect the asteroid shape model using
the spacecraft positions given by the LIDAR processing.
12. Predicted minimum and maximum longitude and latitude of the up to
nine vectors within the NIRS FOV that intersect the asteroid shape model
using the spacecraft positions given by the LIDAR processing.
13. Number of vectors within the NIRS FOV (split into a 3x3 grid) that
intersects the asteroid shape model using the spacecraft positions given by
the LIDAR processing. This number of vectors was used in the calculation to
estimate the previous mean incidence, emission and phase of NIRS, and the
mean, minimum and maximum longitude and latitude values of NIRS.
Processing
==========
The CDR incorporates the best orbital solutions and LIDAR boresight
locations derived by the HAYABUSA LIDAR team. As a first step, a new
algorithm was developed to better locate the Hayabusa spacecraft relative to
the asteroid. The most important data initially used was the housekeeping
(HKEDR) data of the x-y pixel of the illuminated centroid obtained by the
ONC-W1 (WAC) camera of AMICA. Additional data included the project supplied
information on the pointing of the WAC (SPICE C-kernels), as well as a good
shape model of Itokawa (generated by R. Gaskell and part of the HAYABUSA PDS
delivery [GASKELLETAL2008B]). Our algorithm assumes that the spacecraft
attitude (i.e., its pointing) provided by the SPICE C-kernels as determined
by the on board star cameras remained correct throughout the mission.
The algorithm consists of first using a preliminary spacecraft location,
the spacecraft attitude data and the shape model to create simulated images
of Itokawa as seen by the WAC at the time the actual HKEDR was acquired. A
predicted x-y pixel location for the illuminated centroid was computed from
these simulated images simultaneously with a predicted range to where the
LIDAR was pointing at the surface of the Itokawa. These predicted HK-data
were then compared to the actual HKEDR in order to correct the spacecraft
position. This comparison was repeated iteratively until the predicted and
actual x-y pixel locations were within 0.1 pixel, and the predicted and
measured ranges were within 0.5 to 3 m of each other, depending on the range
of the spacecraft relative to the surface of the asteroid.
The algorithm used to reproduce the HK data provides at 2 min intervals
excellent estimates of the spacecraft position relative to Itokawa for most
of the time that Hayabusa observed Itokawa. The data acquired by the LIDAR,
however, was taken at 1 s intervals. Therefore, good estimates of the
spacecraft position were still required for those periods between when HK
data was acquired. After some trial and error, good estimates were obtained
by initially using linear interpolation to first guess the locations of the
spacecraft between those estimates provided by the HK data. We then fit all
the positions using least squares to a second order polynomial or parabolic
function between spacecraft maneuvers. Such a function should have a form
that compares favorably with solutions to the semi-orbital equation of
motion for the Hayabusa spacecraft that include the solar pressure acting on
the spacecraft, because major maneuvers occurred fairly frequently (between
a few hour to a few day intervals). Analysis of the resulting data indicate
significant improvements on how well the new trajectory estimate for
Hayabusa relative to what was initially provided by the project in the form
of SPICE SP-Kernels.
Further processing was then performed, where the following algorithm was
used to improve the unfiltered lidar data. First, the lidar points were
divided into subsets of no more than 1000 points each. The actual number or
points per set was chosen so that the size of a bounding box containing the
points did not completely wrap around the asteroid. Then for each lidar
point in the set, a point on the asteroid near it was computed by
intersecting with the asteroid (using the highest resolution shape model of
Itokawa produced by Bob Gaskell) a ray originating from the spacecraft
position in the direction of the lidar point. These intersection points were
grouped together to form a second set of points in addition to the original
(uncorrected) lidar points. A point matching scheme was then employed to
find the optimal translation of the first set of points so that the distance
between the lidar footprints and shapemodel were minimized. This optimal
translation was then applied to the original lidar points and spacecraft
positions to produce the improved data. This procedure was repeated for each
subset of the lidar data.
In this delivery, we provide three CDR files for each time range. The
first includes boresight locations that have not been further filtered after
the first set of above processing prior to optimization was undertaken. The
second set filters the unfiltered version of the data to remove bad data in
the filtered set: any estimated surface point which was located more that
10m from the predicted intersection of the vector defining the pointing of
the LIDAR boresight and the shape model of the asteroid were removed. This
difference of 10m was chosen because most small scale variations in surface
topography on Itokawa are less than this amount. This second set results in
~65% of the LIDAR points being useful. This is equal to ~1.0 million LIDAR
shots. The third optimized data set was able to use most of the unfiltered
data to further increases the number of usable LIDAR points to ~1.3 million
LIDAR shots. The EDR contains a total ~1.6 million LIDAR data originally
collected by Hayabusa. The optimized data also provided some of the best
estimates of the location of the Hayabusa spacecraft relative to the
asteroid.
Ancillary Data
==============
As part of analysis, we found that the AMICA SPICE image kernel needed
further modification. A new version (amica_v202.ti) was prepared by Olivier
Barnouin-Jha. In order to generate the CDR dataset, we used several project
provided SPICE kernels including the planetary ephemeris kernel
pck00008.tpc, the Itokawa Ephmeris and rotation kernel sb_25143_140.bsp and
the HAYABUSA clock kernel hayabusa.tsc (the version of 2005-09-06). These
and the other SPICE kernels used to prepare the data files have been
archived in the PDS SPICE archives.
Coordinate System
=================
A planetocentric coordinate system is employed, which is body-centered,
using the center-of-figure as the origin. The actual vector from the center
of Itokawa to the surface should be primarily employed for scientific
purposes because of the important curvature of Itokawa where some locations
can possess more than one latitude and longitude. However, latitude and
longitude data are also provided, but should be used with caution. The
latitude is defined by the angle between the equatorial plane and a vector
extending from the origin of the coordinate system to the relevant point on
the surface. Latitude is measured from -90 degrees at the south pole to +90
degrees at the north pole. Longitude extends from 0 to 360 degrees, with
values increasing eastward (i.e., it is a right-handed coordinate system)
from the prime meridian. This coordinate system is preferred for use in
navigation and geophysical studies in which, for example, estimates of
elevation or gravitational potential are generated mathematically.
Modification History
====================
The Hayabusa LIDAR CDR files were changed slightly in going from version 1
to version 2 of the Hayabusa LIDAR data. The dates in the unfiltered and
filtered data filename's were changed to match the dates used in the EDR
filenames, with the exception of the data collected during the last TOUCH
DOWN phase. The LIDAR data collected after November 18, 2005 could not be
corrected. Furthermore, some duplicate data were present in the original
version 1 of the LIDAR data, and these have been removed. Finally, the new
optimized version of the CDR were generated in going from version 1 to
version 2 of the LIDAR CDR. There were also some typos in the version 1
'dataset.cat' document that have now been remedied.
References
==========
Mukai, T., Araki, H., Mizuno, T., Hatanaka, N., Nakamura, A.M., Kamei,
A., Nakayama, H., Cheng, A., 2002. Detection of mass, shape and surface
roughness of target asteroid of MUSES-C by LIDAR. Adv. Space Res. 29,
1231-1235.
Barnouin-Jha, O., A. Cheng, T. Mukai, S. Abe, H. Naru, R. Nakamura, R.W.
Gaskell, J. Saito, and B.E. Clark 2008. Small-scale topography of 25143
Itokawa from the Hayabusa laser altimeter. Icarus 198, 108-124.
Cheng, A.F., O. Barnouin-Jha, L. Prockter, M. T. Zuber, G. Neumann, D.
E. Smith, J. Garvin, M. Robinson, J. Veverka, and P. Thomas, Small-scale
topography of 433 Eros from laser altimetry and imaging. Icarus 155, 51-74
2002.
Gaskell, R., Saito, J., Ishiguro, M., Kubota, T., Hashimoto, T.,
Hirata, N., Abe, S., Barnouin-Jha, O., and Scheeres, D., Gaskell Itokawa
Shape Model V1.0. HAY-A-AMICA-5-ITOKAWASHAPE-V1.0. NASA Planetary Data
System, 2008.
|