Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!gatech!howland.reston.ans.net!ix.netcom.com!netcom.com!nagle
From: nagle@netcom.com (John Nagle)
Subject: Re: Radar on a Chip?
Message-ID: <nagleD5rsJG.n3q@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <D5GM64.BEs@mv.mv.com> <3kils7$fsu@st-james.comp.vuw.ac.nz>
Date: Tue, 21 Mar 1995 03:04:27 GMT
Lines: 27
Sender: nagle@netcom18.netcom.com

empty@mu.sans.vuw.ac.nz (Malcolm Taylor) writes:
>John D. Cook (jdcook@mv.mv.com) wrote:
>: Whats this I hear, in Popular Science no less, about somone coming up 
>: with a micropower radar thats small enough to be fabbed on one silicon chip?

>: If its really real, it sounds great.  The ideal sensor for small, mobile 
>: robots.  Anyone know anything more?

      If you look at the picture on the cover of Mechanics Illustrated,
it appears to be a microprocessor, not a radar chip.  You can make out
the cache and register areas, and notice the large number of pins.
From what I hear, the work is real, but the Mechanics Illustrated
article is a bit bogus.  You can't actually buy the rebar finder they
mention; I tried.

>Ideal? Radar is ideal if you want to scan very large areas. For robots in
>small environments radar will probably not have the required accuracy. It
>would not have much greater accuracy than GPS esp. since it is a small
>silicon built version. If you look at the principles of radar operation
>you will find that to get a 1 meter accuracy you would need approx 300MHz
>timing, (3ns response time) not that easy on a silicon chip. 

      Actually, straight counting at 300MHz is not particularly hard.
You can count to 3GHz with low-cost parts.  Besides, radar isn't done
that way.

					John Nagle
