Why not four sensors each at a compass point on a circle, facing outwards,
with fov 90 degrees each. This guarantees that there are at least two
sensors seeing the beam in each sweep, so you have a guaranteed time
between 'sightings' to calulate distance with.

But I think that using a calibrated power transmitter (like a usual ir
laser diode with feedback photodiode built-in) and simply measuring the
peak pulse amplitude will give distance better for such an arrangement. It
would only need one sensor, omnidirectional. The beam can be coded so it
forms radials (see VOR & TACAN). F.ex. if a radial code is 10 bits
(including crc or better fec) sent in ir with 1msec bit cells and 2 bit
gaps, and each radial is 2 degrees wide, then the beacon rotation time
would be 12*180 msec = 2.16 seconds (27.(7) rpm). This looks like a
perfect speed to implement with a floppy disk spindle motor f.ex. . Each
receiver could then output radial and bearing to beacon after processing.
Using three sensors in an isosceles triangle would also yield receiver
heading wrt beacon 'north' while standing still. This can be refined
essentially forever imho but technical practical considerations make me
think that you can't expect more accurate than 2 degrees of resolution
from this barring a truly professional implementation and very high
quality optics. Cheap diffusors 'lie' about the peak and refract on edges
and molding imperfections like there's no tomorrow (I tried to use such
items as window for ir camera - forget it).

Peter

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body