Newsgroups: comp.robotics.misc
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!lerc.nasa.gov!magnus.acs.ohio-state.edu!math.ohio-state.edu!cs.utexas.edu!news.sprintlink.net!noc.netcom.net!netcom.com!nagle
From: nagle@netcom.com (John Nagle)
Subject: Re: Tilt & De-Acceleration
Message-ID: <nagleDAGAp8.L2x@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <9506181758.AA0015o@hinwick.demon.co.uk> <3s2nop$lae@news.tamu.edu> <9506192138.AA00163@hinwick.demon.co.uk>
Date: Tue, 20 Jun 1995 02:53:32 GMT
Lines: 23
Sender: nagle@netcom6.netcom.com

Mark Worsdall <jaydee@hinwick.demon.co.uk> writes:
>Wesley Dane Scott (wscott@tam2000.tamu.edu) wrote:
>: In article <9506181758.AA0015o@hinwick.demon.co.uk>,
>: Mark Worsdall  <jaydee@hinwick.demon.co.uk> wrote:
>: >Does anyone know of a cheap and simple way of
>: >detecting the difference between TILT and
>: >DEACCELERATION?

      It's a hard problem.  See "Inertial Navigation Systems for
Mobile Robots", in IEEE Trans. Robotics and Automation, V11, #3, June 1995.

      To do this right, you need accelerometers and rate gyros.  
Murata GyroStar rate gyros are around $300 each, and the new
Analog Devices ADXL05 series accelerometers are around $15 each.
You get the high-frequency component from the rate gyros and the low
frequency component from the acceleration vector.  Or you can buy
the whole six-DOF set of sensors in one can from Systron-Donner, but
it costs more.

      Forget liquid-type tilt sensors; accelerometers are now cheaper
and better.

					John Nagle
