Re: Invention Time
Cal asked "How would a CCD correct an unfocused image?"
Sorry, I thought that part was the trivial implementation phase. :-)
Ideas are a dime-a-dozen...
The Adaptive Optics is integrated into the sensor box.
The light might bounce off an array of micro-mirrors (an existing technology: TI's DLP). The exact mechanism would probably need redesign to accomplish this very different function, more up and down, less flipping. Obviously feasible as proven by DLP.
The light might pass through, or bounce off of, a piezoelectric crystal which is laced with a grid of electrodes. Voltages applied would cause a deformation map over the crystal.
The sensor pixels might be installed on MEMS mechanisms to move them up and down. Seems overly tricky.
There might be a small 'wafer thin' flexible mirror that would be deformed by static charge using an array of electrodes. Same basic idea as they use now, but miniaturized.
Remember the Hubble corrective optics? The corrective lenses were the size of coins, for Hubble's 2.4m flawed mirror. Miniaturization and corrective optics work well together.
I can go on all day...
The main point was that 1) yes it's non-trivial, but 2) the performance advantages, cost savings, and logistical advantages are obviously vast.
There's a lot of money spent on Adaptive Optics, so the possible savings might pay for this. Big telescope organizations should take this on, joint venture.