| | | | | I am working on a model that requires me being able to send 6 different
inputs into my program (each will then trigger a motor in either forward or
reverse for a set amount of time). I might be able to get away with only 4
inputs if I get rid of one of the motors (I have some ideas on how to do
that) but I still gotta find an easy way to get the inputs in there in the
first place...
| | | | | | | | | | | | | In lugnet.robotics, Jonathan Wilson writes:
> I am working on a model that requires me being able to send 6 different
> inputs into my program (each will then trigger a motor in either forward or
> reverse for a set amount of time). I might be able to get away with only 4
> inputs if I get rid of one of the motors (I have some ideas on how to do
> that) but I still gotta find an easy way to get the inputs in there in the
> first place...
Rotation sensor? That seems to be the easiest way to record motion, if that's
what you're trying to do. I bet you could probably make a pretty interesting
joystick using a couple of rotation sensors, too.
Adrian
--
http://www.brickfrenzy.com
| | | | | | | | | | | | | | |
| |
| With 3 touch sensors = 3 "bits" = 1+7 possible states/combinations
000 = none pressed
001 = 1 pressed
010 = 2 pressed
011 = 1+2 pressed
100 = 3 pressed
101 = 1+3 pressed
110 = 2+3 pressed
111 = 1+2+3 pressed
or you can use a rotation/angle sensor as a dial (to choose the action) and
a touch sensor as the OK or ENTER "button". ...or you mimic the old
telephone dials, where you reach the number/action you want and the ENTER is
done by returning the dial to 0(zero)/no action.
mc.
> -----Original Message-----
> From: news-gateway@lugnet.com
> [mailto:news-gateway@lugnet.com]On Behalf
> Of Jonathan Wilson
> Sent: sexta-feira, 11 de Abril de 2003 13:09
> To: lego-robotics@crynwr.com
> Subject: how can I get "manual" input into my RCX?
>
>
> I am working on a model that requires me being able to send 6
> different
> inputs into my program (each will then trigger a motor in
> either forward or
> reverse for a set amount of time). I might be able to get
> away with only 4
> inputs if I get rid of one of the motors (I have some ideas
> on how to do
> that) but I still gotta find an easy way to get the inputs in
> there in the
> first place...
>
| | | | | | | | | | | | | | | | | | Marco Correia's utterance expressed in
news:000d01c30037$06378910$610b010a@soporcel.pt:
> or you can use a rotation/angle sensor as a dial (to choose the
> action) and a touch sensor as the OK or ENTER "button". ...or you
> mimic the old telephone dials, where you reach the number/action you
> want and the ENTER is done by returning the dial to 0(zero)/no action.
What a great idea! why haven't I tried that yet?? Fun
things to do with a rotation sensor :) I mainly use it
for distance measuring and crane-boom-control.
Shame on you! Now I'm gonna go to bed too late tonight!
Leg godt!
--
| lars gjerløw jørgensen | lgj[at]jyde[dot]dk |
| N55 43.184 E12 32.405 | www.lgj.dk | oz2lgj |
Mark Hamill:"I have a sneaking suspicion that if there were
a way to make movies without actors, George (Lucas) would do it."
| | | | | | | | | | | | | | | | | | | |
| |
| "Lars Gjerløw Jørgensen" <lgjPURGE@jyde.dk> writes:
> What a great idea! why haven't I tried that yet?? Fun
> things to do with a rotation sensor :)
You don't even need a rotation sensor. You could turn a dial, and for
each position a touch sensor is pushed. Ie, for a typical clock
style, you gear things up so that for one rotation of the dial the
touch sensor is pushed 12 times. A ratchet so that the dial turns one
direction only will be helpful. Then a second touch sensor is used as
the "enter" button.
When I was in high school, back when small towns didn't have
computers, a teacher had a simple instruction kit about computation.
It used almost solely relays that could be wired up to form ANDs and
ORs; there were no ICs, transistors, or other "smart" components. One
project on it that completely amazed me at the time as a simple adder.
The input was a dial, and at each position a circuit would close. You
entered two numbers, pushing a button between each. Then the amazing
thing was that the dial would automatically rotate itself with a motor
until it pointed at the correct answer (ie, the circult would close
and turn off the motor). The same effect could be done in Lego, using
the same input dial as the output.
--
Darin Johnson
Gravity is a harsh mistress -- The Tick
| | | | | | | | | | | | | | | | | | | |
| |
| Darin Johnson wrote:
>
> You don't even need a rotation sensor. You could turn a dial, and for
> each position a touch sensor is pushed.
That's a clever idea. For those who don't want to build such a
device, you can get away with even less: two buttons (i.e., two touch
sensors).
You can model the interface after the cascading menus that many
electronic devices have. Basically, one button is a "next" button
that is used to cycle through a list of choices. The second buttom
is an "enter" button to select a choice. Use cascading menus to
manage complexity. For example, at the first level, use "next"
to cycle through the available motors, and select one with "enter".
Then cycle through the available motor settings (say, forward,
reverse, off) and select one with "enter". Most such menu systems
typically have an exit/back to previous level entry as well. Many
people go through their cell phone menus hardly ever using more than
two buttons (though most provide four directional keys for simpler
navigation).
This is basically just a fancy serial interface for humans, rather
than a parallel one involving many more controls. As a result, it
sacrifices time required to input a command for number of controls
(bits of input).
Interestingly, this is exactly the way you cycle through programs
and select one for running on the RCX itself ;-).
It probably goes without saying that you'd need reasonable feedback
to the user for this to be manageable, although simply changing
the numbers on the RCX's LCD should be enough (digits before
the : might be motor number, digits after : represent direction/
command, etc). Of course, you'd need that kind of feedback to use a
dial too. If you're willing to use brickOS, you don't even need to use
sensors for this, since you can directly control the LCD and also read
from the existing control buttons. The dial idea is still pretty
interesting, though! Basically, it is a rotary attachment to the
"next" button so you can press it many times in succession quickly.
-- Steve
--
Stephen Edwards 604 McBryde Hall Dept. of Computer Science
e-mail : edwards@cs.vt.edu U.S. mail: Virginia Tech (VPI&SU)
office phone: (540)-231-5723 Blacksburg, VA 24061
-------------------------------------------------------------------------------
| | | | | | | | | | | | | | | | | | | | lego-robotics@crynwr.com (Stephen Edwards) wrote:
> Darin Johnson wrote:
> >
> > You don't even need a rotation sensor. You could turn a dial, and
> > for each position a touch sensor is pushed.
>
> That's a clever idea. For those who don't want to build such a
> device, you can get away with even less: two buttons (i.e., two touch
> sensors).
>
> You can model the interface after the cascading menus that many
> electronic devices have. Basically, one button is a "next" button
> that is used to cycle through a list of choices. The second buttom
> is an "enter" button to select a choice.
It works with a single touch sensor as well: use single clicks for the
"next" function, and double clicks to mean "enter". The method is
described in my book too, when talking about the Chess Visual Interface.
Ciao
Mario
| | | | | | | | | | | | | | | | You could also use the Mindstorms Remote Control, which can send Messages 1, 2
and 3. As with touch sensors, you need to use a combination of messages to create
6 separate inputs. (I recall an old thread on this exploring many encoding schemes.)
Jonathan Wilson wrote:
> I am working on a model that requires me being able to send 6 different
> inputs into my program
| | | | | | |