Issues when beginning to work with ranging routine

[Lauren Corlies]

Hello,

My name is Lauren Corlies and I’m an astronomer on the Education and Public Outreach team of the Vera C. Rubin Observatory. Our team is creating solar system activities for students and I am trying to use your code as a way to get the data I need.  In particular, I am interested in using the ranging routine to produce a sample of orbits that represent the uncertainty in fits to different sets of data. 

I’ve downloaded astrometry for 2008 TC3 directly from MPC so I am assuming that it is in the correct format for the code. However, when I run the following command, I get this error:

oorb  --task=ranging --obs-in=2008_TC3.mpc --conf=/Users/dalek/anaconda2/envs/py3/etc/oorb.conf 

  WARNING: NOT using dchi2 in acceptance!

  WARNING: REGULARIZATION is OFF!

***ERROR***  8 Jun 2020 21:43:28UTC (StochasticOrbit / statisticalRanging) No sample orbits found!

***ERROR***  8 Jun 2020 21:43:28UTC (StochasticOrbit / autoStatisticalRanging) First iteration failed.

***ERROR***  8 Jun 2020 21:43:28UTC (oorb / ranging) Ranging failed:

ID: K08T03C and number of observations:  883

I’ve attached the input file I give as well as the config file that came as part of my conda install. 

Is there additional information I am supposed to provide to help it set the initial orbit it tries to fit? What format would that take? 

Any advice you can provide would be incredibly helpful! 

Thank you! 

Lauren Corlies

———————————————————— 

Dr. Lauren Corlies

Astronomer 

Deputy Head of Education and Public Outreach

Vera C. Rubin Observatory

[Mikael Granvik]

Hi Lauren,

the ranging routine is optimized for cases that have scarce data, say, from 2 to
a few tens of observations. The algorithm essentially tries to estimate the
topocentric distance of the object at the observation dates by typically
starting from a very large range such as 0-100 au. If you feed it
hundreds of observations, especially for an object having a (too) close
encounter with the Earth, the actual topocentric distance range that agrees with
the data is maybe 0.00001 au. Hence most of the Monte Carlo samples will produce
orbital solutions that don't agree with the data and are thus discarded. That's
why the code screams: "No sample orbits found!"

As a remedy I suggest that you feed less data to the code. Say, the first 10
observations or so. You can then continue by adding a larger batch of data, say,
20 observations *AND* the orbital solution from the last successful attempt.
This way the code will automatically compute the most sensible ranges to be
covered by the MC sampler:

oorb --task=ranging --obs-in=obs1-10.mpc --conf=oorb.conf --orb-out=obs1-10.orb
oorb --task=ranging --obs-in=obs1-20.mpc --conf=oorb.conf --orb-in=obs1-10.orb --orb-out=orb1-20.orb
oorb --task=ranging --obs-in=obs1-30.mpc --conf=oorb.conf --orb-in=obs1-20.orb --orb-out=orb1-30.orb
...

Note that this sequence is identical to the sequence that you would want to do
in real time/life. That is, you wouldn't want to wait until you have all
observations before computing an orbit but start immediately when you have
obtained the 2 first observations (ie, noticed that it is a moving object).

Hope this helps!

Cheers,
Mikael


--
Mikael Granvik, Ph.D.

mgra...@iki.fi
http://www.iki.fi/mgranvik
Skype: mgranvik

Academy Research Fellow | Associate Professor
|
Department of Physics | Division of Space Technology
Physicum D312 | Luleå University of Technology
P.O. Box 64 (G. Hällströmin katu 2a) | Box 848 (Rymdcampus, E10)
FI-00014 University of Helsinki | S-981 28 Kiruna
FINLAND | SWEDEN
|
Email: mikael....@helsinki.fi | Email: mikael....@ltu.se
Mobile: +358 (0)50 318 2271 | Mobile: +46 (0)76 120 4016
Office: +358 (0)2 941 50715 | Internal: 7579

> --
> You received this message because you are subscribed to the Google Groups "OOrb" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to oorb+uns...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/oorb/7db62987-4d9e-43cc-b28c-2d490e3df763o%40googlegroups.com.
>
>