Re: Ultrasonic sensor interactions

John Barnes wrote:
> In lugnet.robotics, steve <[email protected]> wrote:
> 
>>Ultrasound travels at about 350 meters per second.  If it's
>>detectable range is (say) 3.5 meters then we only need (theoretically)
>>1/50th of a second to emit a ping and get the echo back at
>>maximum range.  The actual duration of the ultrasound ping should
>>be very short - the sound frequency should be up in the 100kHz to
>>MHz range so a very brief pulse is enough.
> 
> 
> The device operates at 40kHz, the transmit and receive piezo devices are only
> resonant at that frequency.

Wow!  That's a lot lower than I'd have expected - at that frequency
there is a good risk of harmonics from normal sound interfering with it.

> Even though the measurement period is short (time from transmit to time to
> receive) it is necessary to wait quite a while for the sound pulse to finish
> bouncing around the environment before starting a new measurement cycle. 

Right - 350 meters per second is pretty slow by computer standards.

> The actual quality of the received signal is such that it is difficult to do
> more than measure the time to the earliest received pulse. After the arrival of
> this pulse, which has taken the most direct route to the nearest surface within
> its "beam" width, further echos will continue to arrive for quite a while, often
> with intensities higher than the first pulse because they are coming back from
> more highly ultrasound reflective surfaces.

(Not to mention things like concave right-angled corners - which make
excellent ultrasound reflectors - but which 'smear out' the signal.)

> It would thus be hard to emit an "id" string of pulses and do any kind of pulse
> compression (by correlation techniques as radar does) to avoid interference or
> even to detect it.

We're doomed.