Newsgroups: comp.robotics
Path: brunix!uunet!charon.amdahl.com!pacbell.com!tlhouns
From: tlhouns@ns.pacbell.com (Lee Hounshell)
Subject: Re: Cheap, light orientation sensor needed
Message-ID: <1992Nov6.213500.6330@PacBell.COM>
Sender: news@PacBell.COM (Pacific Bell Netnews)
Organization: Pacific Bell
X-Newsreader: TIN [version 1.1 PL6]
References: <1992Nov6.171604.29238@kronos.arc.nasa.gov>
Date: Fri, 6 Nov 1992 21:35:00 GMT
Lines: 35

In article <1992Nov6.165554.16576@PacBell.COM> tlhouns@ns.pacbell.com (Lee Hounshell) writes:
>
>I had an idea for constructing one using "bend" sensors with weights attached
>to one end.  As the robot's orientation would change, the sensors would move
>(because of gravity pulling the weights).  If several of these sensors were
>placed strategically inside a robot's chassis, I suspect that one could calculate
>the robot's orientation fairly accurately.  Any comments?

To clarify my earlier posting:

You would need to place 6 of these "bend" sensors in a small enclosed box, one
in the center of each "wall" of the box.  This box would be contained within
the body of the robot.  For an individual sensor, you wouldn't know exactly
which way the tilt occurred.  But by looking at all 6 of the sensors together,
I think you could calculate orientation, as well as "invertedness."

	- the "bottom-most" sensor would always have the most "bend".. as it would
	  probably form a "U-like" shape.

	- the "top-most" sensor would usually have almost no bend, as it would hang
	  mostly straight down.  This (along with the bottom sensor) let you determine
	  up/down orientation.

	- the "right/left/front/back" sensors would allow you to determine roll
	  and pitch.

	- You'd need an electronic compass mechanism of some sort to determine yaw.
	  Anyone know of one?

Anyone know of a reason why this arrangement *wouldn't* work?
Are "bend" sensors flexible enough for this type of task?
Is there a way to easily attach them "perpendicular" to a wall?
And add a weight to the free end?

-Lee
