Introduction to Robotic Astronomy - Part I

Joe Shuster

If you're like most novice astrophotographers, you reach a point of progress where your sessions have fewer and fewer struggles with alignment, tracking, software, interfaces, cameras, focus, filters and all the other elements that make imaging such a fun exploit. You might still trip on this or that: one night the telescope goto's will be off; another night the computer will run slowly, another night you'll struggle with focus seemingly forever.

Eventually, there's a special night -- A night where you start a long sequence of images after a minimal amount of setup troubles. You realize that the next step -- after 30-60 minutes of automatic exposure -- is to reposition the telescope and start another sequence. For about 10 seconds that's good feeling - that "YES!" feeling. But then you might start wondering what you're going to do next. After all, with no serious technology issues to solve you've been reduced to a night watchman. And this feeling of diminished importance can be amplified by ambient aspects such as frigid temperatures or sweltering heat (punctuated by blood thirsty insects). Worse, as you sit silently staring at the sliding progress bar of the 26th of 45 one-minute exposures you feel like your eyes are getting a little heavy.

You start to think about how you can do all the necessary tasks without sitting outside. You've gotten the "robotic astronomy" bug. You're not the only one. Many people have addressed the situation. The purpose of this article is to give you some tips on getting started with finding your solution.

Let's define some scope - I won't be talking about the public robotic astronomy offered by commercial sites. Sure you can rent optics and electronics at remote sites and via the Internet produce your own photos. If that's your direction, you don't need any additional guidance from me or anyone else. Personally, I liken that kind of sterile, distant operation to having someone else take a vacation to Hawaii and then send you the videos they took. Certainly something seems missing from the process, but that just my opinion.

I also won't be talking about the high-end robotic astronomy with dome control, weather sensors, Internet access, etc. That's all possible but it's way beyond the "introduction" stage.

Robotic astronomy isn't for everyone. If you are a visual astronomer, there's not much point in robotic operations. And if you are a computer illiterate by nature or choice you won't find much use for robotic astronomy. So let's decide what kinds of activities we can discuss within our scope: First, you want to be able to start an imaging process and monitor it. You will also want to be able to aim the optics at your chosen object. You also might want to use an autoguiding system to keep long exposures properly aimed at the target. For now, let's presume your imaging targets are DSO's. (Lunar and planetary imaging has some different characteristics, so I'll mention how that affects your robotic setup later.)

This immediately provides some specifications for equipment. You'll need telescope capable of being automatically pointed - a GOTO scope. You'll need a camera that can be controlled via an electronic connection to the camera for shutter and other camera settings. Finally, you will need a PC for all the kinds of monitoring you'll want to do.

So let's start with the simplest setup that could qualify for "robotic astronomy". You have a telescope capable of basic goto operation and you've aligned the telescope, preferably in an equatorial alignment. (If you're not sure why, browse the Internet for information about "field rotation".) Your camera is attached to the computer with the ability to digitally control the shutter and examine pictures.

Without any additional complexity - hardware or software - you might like to take the telescope hand control and computer inside. That would give you telescope control and camera control. Unfortunately, this process isn't as simple as it initially seems.

The first problem you might encounter is the length of cable between your camera and computer. Let's say you need about 50' between the telescope setup and your indoor location. Most cameras use USB for connections, others use serial or parallel connections. Serial and parallel connections aren't so bad - that length isn't technically recommended but it works. However, USB connections aren't supported longer than 15'. There are "active" USB cables (about 25' long) that boost the signal. These can be chained together in theory. But in practice, a lot of people find that these devices are very sensitive to outside conditions - cold and moisture especially - and the active cables are not cheap. I found that active cables start getting flaky when the temperatures are in the 40s.

There are long-distance USB setups using Ethernet cabling. Using one of these, for less than $100, you can be operating long-distance USB 1.1 (close to $200 for USB 2.0) at a range of about 150' (with some concessions). Typically, these devices have a single USB female connection in a control box at the remote location and another control box at the controlling location that plugs into the computer's USB port. The control boxes are connected via the cables used for Ethernet communication. (Cat5 cable with RJ45 connectors.)

Another issue is with the hand controllers of most goto telescopes. Most hand controllers can be wired with an electronically compatible extension cable or you can entirely replace the original cable with a longer cable. Several third parties offer extension cables for this kind of use. Some people try to get "remote control" by adding a long extension cable between the controller and the telescope. But once again, in practice, these long cables are frequently the source of erratic behavior. Even more seriously, "hot plugging" hand controller cables (by intention or accident) can cause serious electronic damage to the mount.

So for all practical purposes, expect that your hand controller will be limited to the immediate vicinity of the telescope. That means you'll need a scope that can operate over a serial connection so you can aim the mount at the various objects you want to capture. Most contemporary scopes have that capability, but you'll need a cable that goes from the mount or hand controller to a computer. Getting the serial signal to a distant computer isn't technically challenging. You can use a long serial cable or you can use a serial to USB interface and plug the USB into the computer.

Meanwhile, there's another downside to the idea of stretching a set of long cables from the telescope site to the computer site. It might start simply with a long cable for the camera and a long cable for the telescope. But if you decide to add things like a remote control focuser, motorized filter wheel, autoguiding camera or anything else, you'll need a proper distance cable for all those things. So in addition to the unavoidable cable jumble at the telescope, you'll need to trail all these cables inside. Basically, the long cable philosophy doesn't scale well.

There is a slightly more sophisticated (and more practical) approach to using "long cables". (Let's call that "Stage 1" robotic astronomy). There are products that extend your keyboard, mouse and video over a cable. These are called KVM cables. With these cables, you can set up a work area with a mouse, monitor and keyboard to control the computer at the telescope (described later). KVM cables run about $60 for 30' length. That's not a significant distance for most people. You can accomplish longer runs with KVM extenders. A basic extender has two sets of mouse, keyboard and video connectors - one for the computer and one for the devices. These extension adapters then can be connected by a "CAT5" cable for lengths up to 200'+. A typical setup of a KVM extender and 50' cable would run about $200. Of course you'd also need the keyboard, mouse and monitor. Also, you'd need to be sure that the computer at the telescope site could easily switch control between the built-in mouse, keyboard and monitor to the inside devices. Some laptops aren't keen to switch the keyboard or mouse that easily.

I'm sure that people are using the Stage 1 method (either variation) for robotic astronomy, but I cannot personally encourage it because of its limitations and poor scaling.

But there's a variant of the Stage 1 approach that has some attractiveness. Most goto telescopes have the ability to connect to a computer using a USB or serial cable so you can control and monitor the telescope independent of the hand controller. You need some control software - a planner program or planetarium program or even some imaging programs - that is capable of talking to your telescope. The telescope interface cables are inexpensive and the software ranges from free to over $100 or more. With such a setup and some simple USB interfaces, you can safely move your computer - maintaining camera and telescope control -- about 20' away from the location of the telescope.

What can you accomplish with a 20' range? Well, at home it doesn't help much. If you're 20' from a home desktop computer you'll be lucky to be outside! And even then you'd face some "horizon" challenges to say the least because your house would probably block 50% of the sky. But consider an imaging setup at a remote dark site. 20' might be fine to give you the range to slip inside a comfortable tent or into your vehicle. Of course, you'd need a laptop and sufficient power to operate the whole setup for quite a while. But you'll have the benefit of being in an enclosed space - whether it's warmer or just more "bug free".

So we've started to explore the requirements for robotic astronomy and some of the problems that crop of as we try to develop the environment. In part two, I'll go into more detail about truly networked remote operations, including an list of some resources for the products I use as well as some general resources for setting up a remote astronomy environment.

Published in the April 2006 issue of the NightTimes