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: Bicycle Lean Angle Sensor
Message-ID: <nagleD6I888.3np@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <3lek78$p8n@bellboy.ucc.uconn.edu>    <wpenrose.13.00134AF5@interaccess.com> <3lpj1e$uch@newsgate.sps.mot.com>
Date: Tue, 4 Apr 1995 09:40:56 GMT
Lines: 28
Sender: nagle@netcom20.netcom.com


>wpenrose@interaccess.com (William R. Penrose) writes:

>> >I know that a gyro would solve my problem, 


     Well, not necessarily.

      First, you have to decide whether you want the gravity
vector or the current inertial "down" vector.  For balancing
a bicycle, the latter is probably sufficient, although roll rate
may be necessary.

     The trouble with pendulums is that they oscillate.  You can
fix this with damping, but that just applies a low-pass filter to
the output, and your information is always late.  What you really
want is a pair of solid-state accelerometers.  This will give
you the instantaneous inertial "down" vector, and if you servo the
bike to keep the horizontal component (in the bike's reference frame)
zero, that's probably enough.  You may only need one accelerometer,
in fact, mounted to detect the X component in the roll plane.

      These things are cheap (< $100 US, maybe less.)  Look in the
FAQ for suppliers.

					John Nagle


