Re: [find_orb] Re: Getting constrained orbits

Bill J Gray Jul 20, 2010

Hi Diego,

My apologies; that message slipped by me!

As you mention, getting positions from two or more observers widely
separated will help to refine the distance to the object nicely. This
will happen even if they aren't observing simultaneously. In that
situation, there's no need to impose a constraint on R1 or R2. Those
constraints will just fall out, mathematically, from the observations.

To put it another way: suppose two widely separated stations are
observing an object at roughly the same time. The path shown from one
station is about 12" away from that shown by the other observing
station.

In this situation, when Find_Orb tries out various orbits, it
will find that only those with a narrow range of R1 and R2 produce
reasonable residual errors. So it may produce "wrong" orbits, but
any orbit it produces will have roughly the correct value for the
distance to the object, and there is no need to explicitly constrain
R1 and/or R2.

This is a very common situation. If an object appears on NEOCP
and has been observed from two widely separated stations, the
distance to the object is (usually) nicely defined, unless it's
very far from us. The only thing that isn't well-defined in
such a case is the radial velocity; it takes many more observations
(usually) to figure that one out.

This list _has_ been quiet, perhaps in part because Find_Orb
hasn't been updated lately, not since the 2010 April 10 version at

http://home.gwi.net/~pluto/devel/find_orb.htm

I do have a version almost ready that adds the ability to compute
comet non-gravitational parameters (A1 and A2), and has a button for
the "downhill simplex" method described at

http://www.projectpluto.com/herget.htm#simplex

These features are working, but I've one more I want to get
completely fixed: statistical ranging. This will allow one to
compute Monte Carlo type orbits even for very short arcs, and
even in the case where one has only two observations. Doing this
hasn't actually been all that tough, but I want a version that
will be very robust and won't require the user to know too much
about orbital mechanics. I'm just about there.

-- Bill

Diego R wrote:
> This group seems to be on holidays!
>
> Well, thanks to it, I could see my question was already answered at http://www.projectpluto.com/find_orb.htm#Convergence_problems
>
> But I have yet to figure out how to get R1 and R2 from a single distance at a given date.
>
> Thanks for any hint!
>
> Diego
>
> --- In find_orb@yahoogroups.com, "Diego R" <incorregible_98@...> wrote:
>> One of the nice features of Find_Orb is its ability to self impose contrains in one or more of the six orbital parameters. But sometimes, what you got to offer Find_Orb in order to make things easier is distance. Distance to Solar system minor planets can be determined via parallax using two observers in the surface of Earth imaging the object at the same time.
>>
>> For instance, two observers 10 000 km apart will get a noticeable 12,50" parallax from an object 1,10 AU away.
>>
>> How could that distance be used to help Find_Orb refine its orbit calculations?
>>
>> Thanks!
>>
>
>
>
>
> ------------------------------------
>
> Yahoo! Groups Links
>
>
>
>