Subject:
|
Re: Help with my robot
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 12 Dec 2000 16:34:03 GMT
|
Original-From:
|
AGH <JEDI_AGH@nospamYAHOO.COM>
|
Viewed:
|
558 times
|
| |
|
|
Thanks for your input, need to rethink what I am going
to do...
--- Steve Baker <sjbaker1@airmail.net> wrote:
> Ahui Herrera wrote:
> >
> > I got a problem. I have a simple two motor robot • that I want to have a push
> > button cause the following:
> >
> > 1 hit the push button
> > 2 robot turns 180 deg and move forward
> > 3. after 15 feet robot stops
> > 4 repeat 1 - 3
> >
> > The problme i am having is that the robot will not • go straigt. it is not
> > turning 180 degrees.
>
> As lots of other people have said - you really need
> rotation sensors.
>
> If you don't have one - then a line-follower is an
> option (as other
> people have also suggested)...or I *suppose* you
> could put a bright light
> at either end of your 15 foot track and have the
> robot aim for the brightest
> light source after it's done an approximate 180
> turn. If you go this way,
> you'll want the smallest (but brightest) light you
> can find.
>
> But to do this properly - you'd certainly be better
> off with rotation
> sensors - if you can do this - you should.
>
> I suppose you could try to make your own rotation
> sensor by taping
> a black/white strip of paper around a large wheel
> and use the light
> sensor to count the number of stripes that go by.
> This scheme makes
> detecting the DIRECTION of travel rather difficult
> though (not
> impossible though - we can talk some more if you
> want to go this
> way).
>
> However, in your case, we don't need a full-blown
> rotation sensor.
>
> For steering, you could arrange to have just one
> narrow black strip
> on the detector wheel and use a differential
> connected to the two drive
> wheels to rotate our detector wheel. Then, when the
> robot is going
> straight, the black mark will remain under the light
> sensor. If the
> robot veers off to the left or right, the black
> stripe will no longer
> be under the light sensor. Unfortunately, you won't
> know whether to
> turn left or right to correct for this - so you may
> have to turn left
> for a little while - and if you don't see the black
> stripe, turn right
> until you see it. The software to do that turns out
> to be IDENTICAL
> to the code for a line-following robot!
>
> In a sense, the robot is following a line that it
> carries along with it!
>
> You could use the same technique to turn 180 degrees
> accurately by
> arranging the gearing such that the detector wheels
> does an exact
> whole number of rotations after a 180 turn. Then
> all your software
> would need to be would be:
>
> * If the bump sensor hits, turn for X
> seconds...which should
> be approximately right for a 180 degree turn.
> * Now, if we got it right, the black strip on the
> detector
> wheel will be under the light sensor again.
> * Run a line-following algorithm until the bump
> sensor
> hits again.
>
> Your description of the problem said you had to
> drive 15 feet.
> That's rather hard to measure when you are doing
> 'line following'
> because the robot has to stop frequently to find the
> line again,
> and when you do that, you tend to lose track of
> distance travelled.
>
> However, if what you really mean is to run until you
> hit the bump
> sensor again (which happens to be a distance of 15
> feet) then
> what I suggest could work.
>
> Good Luck!
> ---
> Steve Baker HomeEmail: <sjbaker1@airmail.net>
> WorkEmail: <sjbaker@link.com>
> HomePage :
> http://web2.airmail.net/sjbaker1
> Projects : http://plib.sourceforge.net
>
> http://tuxaqfh.sourceforge.net
>
> http://tuxkart.sourceforge.net
>
> http://prettypoly.sourceforge.net
>
>
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
