From: dstamp@watserv1.waterloo.edu (Dave Stampe-Psy+Eng)
Subject: Re: head tracking with head-mounted lasers
Date: Wed, 27 Nov 1991 18:44:10 GMT
Message-ID: <1991Nov27.184410.15491@watserv1.waterloo.edu>
Organization: University of Waterloo


70353.3056@CompuServe.COM (Christopher Fry) writes:

>Eric Drexler showed me his laser pointing device that he uses when lecturing.
>It is like a small flashlight, 1 inch in diameter and about 6 inches long. The
>weight is essentially that volume worth of batteries. It lasts 6 hours on 
>them. Cost $200. It puts a small, intense red spot on whatever surface it's 
>pointed at.
>
>So put 3 of them on your head and point them all towards the ceiling at 
>different angles. If you restrict movement to a 4 ft x 4 ft area, and you 
>cover that area with a 525 line videocamera [aimed at the ceiling], you get a 
>resolution of about .1 inch. If that's not good enough, do the predictive 
>tricks mentioned previously to interpolate points.  
>
>4x4 ft isn't big enough, of course, but its about what you get from the 
>polhemus. Need more room? Just add cameras.
>
>You could get 8x8 with 4 cameras and keep them at the periphery ok, but more 
>would be a problem as they'd interfere with you and/or the laser beams.
>
>Run some power to your head and you can cut out the battery weight. Keeping 
>the cameras off your head will be cheaper and lighter. 

Ummm...  Problems (2 of them):

1) This would have to be done in a darkened room, as the laser spot is
in the visible light range.  You can use IR in a room with fluorescent
lighting.  Plus you need a smooth, matte reflective ceiling.

2) head angle (up/down, roll from shoulder to shoulder) will cause large
motions of the spots, well outside the 4x4 foot square.  Left-to-right
rotations of the head will have TINY gains by comparison.

Also to be considered is where to put the camera: you don't want to have a 12'
ceiling height.

It's easier to put IR LEDs on your head and look at them directly.  By proper
selection and mounting, you can see the LEDs over +/- 60 deg. of 
rotation, and the spots are TINY.  Problem still is getting good angular
resolution for head rotation.

--------------------------------------------------------------------------
| My life is Hardware,                    |                              | 
| my destiny is Software,                 |         Dave Stampe          |
| my CPU is Wetware...                    |                              | 
| Anybody got a SDB I can borrow?         | dstamp@watserv1.uwaterloo.ca |
__________________________________________________________________________