Re: [Phys-L] parallax in astronomy

[John Denker <jsd@av8n.com>]

On 11/02/2014 02:30 AM, Kauhanen Kauko wrote:
> It helps to understand the method if you apply it in a classroom and
> use common units of distance!

Agreed!

Not everybody is motivated by astronomy.  Just because it
is interesting to me doesn't make it interesting to everybody
else.  There are city dwellers who never see the stars, much
less measure their distance.

Parallax has lots of real-world applications.

Ordinary stereopsis is useful for things like trying to swat
a mosquito ... but people tend to overestimate its power, to
the neglect of useful monocular cues such as /motion parallax/.

For example, suppose you are driving a car at some constant 
speed.  You see some object dead ahead.  It's angular size has 
doubled in the last minute.  That means you will be there in 
less than a minute.
 -- In the small-angle approximation, you will be there
  in exactly one minute.
 -- If its size doubled from 5 degrees to 10 degrees,
  you will be there in 0.98 minutes.
 -- If its size doubled from 30 degrees to 60 degrees,
  you will be there in half a minute.
 -- It its size doubled from 45 degrees to 90 degrees,
  you are already there.

Motion parallax also works perpendicular to the direction of
travel.  Suppose you see some stationary object that is directly 
abeam, at the 3:00 position.  One minute ago it was 5 degrees 
ahead of 3:00.  That's 1/12th of a radian.  If you are traveling
at one mile per minute, the object must be 12 miles away.


Now suppose you are trying to land an airplane.  This involves
judging height above the runway to high accuracy, within a
few inches.  Theory says that stereopsis "should" be usable
at these distances, but in practice monocular cues are vastly
more powerful.  You can land the plane just fine with one
eye closed.  Famous aviator Wiley Post had only one good eye.

You might have a hard time guessing a_priori what are the best
cues.  Every student tries to do it using geometry based on
runway width, but this fails miserably when you go to a new
runway with a different width.  Runway length is worse.  Using
Joe's Cafe as a height-gauge is also a really bad idea.  For
a discussion and diagrams of what actually works, see
  https://www.av8n.com/how/htm/landing.html#sec-flare-timing


There are lots of animals with eyes on the sides of their head,
with no overlap in the fields of view, and hence no stereo depth
perception.  The can however perceive depth, by bobbing their
heads and using motion parallax.