Feature #2221

deltack requires iterative application for edge cases with poor pointing and small targets.

Added by Moses Milazzo almost 4 years ago. Updated 12 days ago.

Target version:
Software Version:
Test Reviewer:


This is an external support board issue.,3852.0.html

It is likely that deltack will require a redesign to fix this issue.

Hello, I'm trying to do a simple deltack on images from the New Horizons Jupiter Encounter.
I run through lorri2isis, do an editlab to swtich TargetName to Io, and then do a spiceinit attach=yes. However, when I try and deltack, I get this:

deltack fr= Io_deltack.cub samp1= 437 line1= 191 lat1= -5.249 lon1= 216.651
starting iteration 1
Iteration: 1
Sigma0: nan
Observations: 2
Constrained Parameters:3
Unknowns: 5
Degrees of Freedom: 0
Deltack Bundle has converged

Generating report files

Bundle complete
Group = "Iteration1: Final"
Sigma0 = nan
Observations = 2
Constrained_Point_Parameters = 3
Constrained_Image_Parameters = 0
Unknown_Parameters = 5
Degrees_of_Freedom = 0
Rejected_Measures = 0
Converged = TRUE
TotalElapsedTime = 5.41e-04
ERROR Unable to update camera pointing for [Io_deltack.cub].
ERROR An unknown NAIF error has been encountered. The short explanation provided by NAIF is [SPICE(NOTAROTATION)]. The Naif error is [Input matrix was not a rotation.].

I hope the SPICE error is more instructive for you than it is for me. The intent is to move the pointing only by about 9 pixels.

In this particular case, Io was only a few pixels across, and that motion of 9 pixels was 'off the disk' of where SPICE thought Io was. I was told by a colleague (Schenk who heard it from Tammy) that deltack only works when the 'motion' is within the disk of the planet that SPICE is knows about. So to deltack your way more than a planetary radius requires several applications of deltack, moving about 2/3 of the disk each time.

So the solution requires me to iteratively apply deltack, when I'd just like to do it in one go.

Related issues

Related to ISIS - Bug #1317: deltack - problems with re-assigning subspacecraft lat/lon for adjusted pointing on flyby data Acknowledged


#1 Updated by Tammy Becker almost 4 years ago

jigsaw encounters the same problem. Every pixel coordinate requires an apriori lat/lon in order to perform a solution.

Paul Schenk has been running deltack first to adjust the pointing onto the disk within an image first in order to run jigsaw.

#2 Updated by Tammy Becker over 3 years ago

  • Status changed from New to Acknowledged

#3 Updated by Moses Milazzo over 3 years ago

  • Priority changed from Low to High

For some reason I had set the triage_priority to 1, which doesn't make any sense at all. Fixed.

Also available in: Atom PDF