Re: Occupation sensor that works with people sitting still?

[mike <ki4fyy@xxxxxxxxx>]

>>A sensor placed in the doorway to the room could 'count' the number of
>>occupants in the room. It would have to detect the direction of travel
>>so that entering the room would add 1, exiting the room would subtract
>>1. When the count is > 0 then there is someone in the room. When the
>>count is 0 then the lights can go off.
>
>
> What if there's more than one exit to the room?  What if the person stood in
> the door, changed their mind an went back into the room?  Entry/exit is too
> imprecise and error prone to be used alone.
>

I think the original post was for use in a bedroom -- most bedrooms only
have one entrance. Even if there were 2, the "in-counts" could be "or'd"
and the "out-counts" could be "or'd". To make it robust, the direction
of travel would need 2 sensors and they could be set apart by maybe 3
feet or so. This might not be practical unless the doorway led to a
hallway. Use modulated IR light beams for the sensors and TV remote
control IR receiver modules to avoid problems with extraneous lights.
I've seen sensors in museums and restrooms that probably work like that
to shut off the lights when nobody is present. My guess is that someone
could do all this with a basic stamp and a handful of other parts.