But... it’s a telescope. Couldn’t it determine direction from images received from a spotting or wide angle lens? Does it not have one?
Each of the Hubble's gyroscopes only measures rotation around a single axis.
> But... it’s a telescope. Couldn’t it determine direction from images received from a spotting or wide angle lens? Does it not have one?
Hubble does have imaging sensors to allow it to track the stars and orient itself. But the orientation needs to be not only measured very accurately (to within a few millionths of a degree), but also kept stable over long periods of time, in order to avoid motion blur. So the attitude control system needs to have low latency.
In principle it would probably be possible to do this using only image sensors, but it would have been very challenging when that Hubble was designed in the 1980s. It has a number of CPUs, the fastest of which is an 80486 running at 25 MHz.
Not unexpected but reading that now... Absolutely remarkable.
Was it installed as a later update? The PowerPC G3 was released in Apple computers in 1997, just a few years later.
[1] https://en.wikipedia.org/wiki/DF-224
It was roughly 45 centimeters (1.48 ft) by 45 centimeters (1.48 ft) by 30 centimeters (0.98 ft), weighed 50 kilograms (110 lb)
It had about 96KB of usable plated-wire memory (32K 24-bit words).
That seems like a lot of weight. I wonder how much was radiation shielding?
The fact that we can even achieve this level of precision from a piece of hardware blows my mind.
This instrument can resolve an angle change of < 0.3 millionths of a degree in one second.
https://arxiv.org/abs/1309.4828
Disclaimer -- am an author.
Edit: and this instrument, using the autocollimator above, can do the same thing for inertial sensing. Same disclaimer applies. https://aip.scitation.org/doi/10.1063/1.4862816
Hmm, I think the second time the gyros needed replacing, and the gyros were the limiting factor on the telescope's lifetime, I'd have installed a redundant set.
But as other mentioned, they had redundant set of gyros...
This is a common fallacy in computer vision. Often people get to a within a pixel and stop because they assume they can't get better. Most times you can get substantially better. I worked on a qr-code like system where the scanner could reconstruct the code from an image with (slightly) less pixels than 'pixels' in the code.
Not obvious a priori, but makes sense.
The light is blurred, with a fairly predictable pattern, so if you fit a function to the shape, you can find the peak of the function and that is the most likely center position for the point source.
There is the star tracker camera (widefield, fast), and the imaging camera (narrowfield, slow).
Star trackers use a fast widefield camera because it's easy to get a good signal:noise ratio (there is a lot of contrast between the stars and the background).
The imaging camera, on the other hand, generally takes much longer exposures (HST subjects are generally very dim, relative to stars). All those beautiful, nebulous HST photographs you see? Those have exposure times on the order of hours or days. In practice, being off by a pixel momentarily is not a big concern -- the amount of "bad" photons you collect during that time is very small.
The gyros are only used for coarse pointing, guide stars are used for much more accurate and precise position sensing during exposures (which is why they are called Fine Guidance Sensors).