Automated Telescopic Tracking of Aerial Objects: Proof of Concept

Since October I've been working on developing a program to enable most standard GOTO-enabled telescopes to track a wide variety of objects in the sky in an automated fashion. One obvious application of such a program would be to enable UFO researchers to set up home observatories capable of tracking unidentified aerial phenomenon in a semi-automated way. Likewise it could greatly simplify tracking known objects as well, including planes, rockets, or even satellites. The program is still at a very early stage of development, but it provides manual joystick control (as an analog input; the greater the joystick deflection and the higher the throttle setting, the faster the telescope will slew) as well as automatic video tracking of objects using an LX200 Classic telescope. It will have other features as well, including video-guided satellite tracking. I have conducted some "proof of concept" tracking with the software, including a combination of manual and automatic tracking of the recent Atlas V OA-4 rocket launch and this weekend I tracked several aircraft on final approach to a local airport.




This weekend I will also be attempting to track the return-to-flight of SpaceX's Falcon 9 rocket, assuming it launches on Saturday or Sunday (and assuming it's not as cloudy as the last launch). This launch promises to be particularly dynamic if they win approval for their planned attempt to land the first stage of the booster at a pad out at Cape Canaveral. In theory I should be able to track not only the launch, but some of the descent burns of the first stage as it makes this historic attempt to propulsively recover a vehicle for re-use after an orbital launch.

SpaceX may also attempt to land the Falcon 9’s first stage at a landing facility the company leases from the U.S. Air Force at Cape Canaveral. If the landing attempt goes ahead — it is pending final regulatory approvals — it would mark a new chapter in SpaceX’s efforts to prove the reusability of the Falcon 9 after being unable to nail landings on an offshore barge.

spaceflightnow.com...

This would be especially useful as a proof of concept for unidentified object tracking; upon first stage re-ignition the first stage of the vehicle will re-appear in the night sky for a very short period of time and although I will be watching for it deliberately I will not know precisely where it will appear beforehand. Unlike tracking planes or rockets launching from a pad, the starting location will only be a general direction (somewhere in the eastern sky from my location) and the time available to observe it will be measured in mere seconds since each re-light of the stage is quite short in duration. Even knowing that it will appear somewhere in the eastern sky, it still presents a very transient and challenging target to capture on video through the high magnification and small field of view of a telescope.

Depending on the exact flight profile it should be possible to observe at least 1 or more re-lights of the first stage. This diagram shows a basic flight profile for the first stage, though the landing location will be back near the launch site rather than a floating barge if approval is granted to attempt it:


i.stack.imgur.com...

Ultimately I hope that this program will prove to be a useful tool for serious UFO researchers who wish to use commercial off the shelf GOTO telescopes to record unidentified aerial phenomena. In developing this program I will need community support in the form of beta testers. The telescope I use is quite powerful but also technologically obsolete (the Meade LX200 Classic - pre-GPS, pre-Autostar). I will be porting the program for use with Autostar and Nexstar controlled telescopes, but I do not have access to those scopes. I need beta testers who own more modern Autostar and Nexstar controlled telescopes and are willing to test this software on their equipment. Unfortunately the software hasn't generated any interested on astronomy forums; not many amateur astronomers are interested in using their expensive telescopes to track aircraft or rockets or things of that nature.

There is a similar (and expensive) program already on the market which fulfills a similar function, but its compatibility list is extremely short when it comes to Meade telescopes. In fact it's exclusively compatible with newer Meade LX200's, but is completely incompatible with older LX200's like mine. I would love to support the majority of Autostar controlled Meade telescopes (as well as older Magellan controlled LX200's like mine of course) as this would greatly widen the potential user base for this software. The problem is the lack of willing participants for beta testing. So far I have not had any volunteers for beta testing, and this may ultimately doom my plans to release to software for use by others. I know it works with the LX200 Classic, but very few people still use that old scope anymore. Most use newer Autostar or Nexstar telescopes. While I can easily port the program and covert the scope communication lines to the formats of the Autostar or Nexstar telescopes, I can't risk distributing completely untested software for obvious reasons. Even with my own older and simpler scope there were unexpected quirks involved in communicating with the telescope which I had to work out on my own through several early versions. The more beta testers I can get representing a wide variety of GOTO telescopes, the better.

a reply to: ngchunter

Sounds very interesting...wish I still had a telescope.

Fascinating!, sad so say mine is a 10" Dob....

A good GOTO telescopes, is on my bucket list for next project

So may be talking to you before I buy
working on my second all sky camera now

originally posted by: manuelram16
Fascinating!, sad so say mine is a 10" Dob....

Well the bad news is there's not much I can do to help you with your dobsonian in terms of tracking, the good news is that the dobsonian mount was designed specifically to make hand tracking easy to do and I have seen people get good results hand tracking satellites with dobsonian telescopes by making sure their viewfinder is well-aligned.

a reply to: ngchunter

Do you know of any guides that one can follow for tracking objects with a dobsonian? I've got a 12" dobsonian and have logged a view views of deep sky objects, but nothing in orbit.

You're system might be good for a slow moving Foo Fighter --- but for a Foo-Fighter in it's hi-power bluish-white phase --- you'll probably need a video camera that can track a lighted object, going at approx. 17,500 mph at around 5000+feet in altitude.

originally posted by: Erno86
You're system might be good for a slow moving Foo Fighter --- but for a Foo-Fighter in it's hi-power bluish-white phase --- you'll probably need a video camera that can track a lighted object, going at approx. 17,500 mph at around 5000+feet in altitude.

Depends on distance and on the telescope specifications (maximum slew speed). That's for the hardware side. I'm simply providing the software, the hardware you select and its maximum slew speed is up to you. Tracking fast moving rockets and satellites will be possible for the majority of telescopes this software is compatible with.

a reply to: ngchunter


Nice!

I'm building the same sort of functionality into my robot. Though not necessarily for tracking fast objects, in fact I didn't even consider anything faster than say a passing asteroid, but after checking on the specs of the mount I'm thinking of, it should be possible to easily track a fast moving object.


I noticed that your system seems to "loose it" from time to time, not sure if it's periodic...didn't check. Also that it didn't keep objects centered in the field of view well, something that is essential if not capturing video. But, you still have an "alpha" version, so I'm sure all that will improve with additional development.

Mostly curious...what language are you using?

originally posted by: tanka418
a reply to: ngchunter
I noticed that your system seems to "loose it" from time to time, not sure if it's periodic...didn't check. Also that it didn't keep objects centered in the field of view well, something that is essential if not capturing video. But, you still have an "alpha" version, so I'm sure all that will improve with additional development.

The video is partly manually tracked, so you shouldn't judge the ability to keep objects centered based on this video alone. Future videos will showcase more of the automated tracking. There may be a SpaceX Falcon 9 night launch this weekend which I will be tracking. The previous launch video I posted above was mostly manually tracked due to extremely bad cloud cover interrupting the automatic tracking almost immediately after the vehicle left the pad. The weather is forecast to be much better this weekend.

When the weather is clear this program is superb at keeping objects centered in the view. The problem is that when it passes through large clouds you have to take back manual control to find it after it emerges from the cloud. You can re-engage automatic control once you find it again but it takes a bit of practice to do so smoothly and quickly when the object is fast-moving like a rocket. With the planes it was again a combination of manual and automatic control due to "distracting objects" in the viewfinder window. I'm working to improve the detection algorithm to intelligently ignore all similar-looking objects in the viewfinder except the one specifically being tracked.

2015 YB is a newly discovered small asteroid passing very close to earth tomorrow. If the weather allows I will track it tonight using this software as well (it features open-loop tracking of asteroids in addition to the closed loop video tracking).
www.minorplanetcenter.net...