Subject:
|
Re: RCX & RIS, a fading glory?
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Thu, 30 Jan 2003 00:47:39 GMT
|
Viewed:
|
1559 times
|
| |
|
|
In lugnet.robotics, Mark Riley writes:
> I see no reason why TLC couldn't just continue to manufacture RIS 2.0. Why
> continue with R&D to develop 3.0 if 2.0 is more than adequate to stimulate
> young minds into considering a career in engineering? 3 sensor inputs and 3
> motor outputs are more than sufficient, given that TLC's perceived market
> for the product is older children (i.e. not the adult variety). In fact,
> limiting the number of I/O ports forces one to be creative in solving a
> problem, instead of just throwing more sensors or motors at the problem.
I agree. One thing I thought is that, Keep the RCX and offer an expanded
version of the Technic Control Center. It currently has three motor
controlls via the pad keys and plays back movments. I'm making a giant robot
that has 6 motors on just one arm. I need a better way to manage all those
motors. I would like to have a Technic MULTIPLE Master Control Center
containing say 12 motor ports. You would plug it into the wall then be able
to power all you motors.
Then ontop of that the TMMCC would have the ablitity to record the movents I
program using the polarity keys, then be able to transfer that set of motor
movments into Code that you could manipulate from within your computer.
That is to say, hit the motors keys to create a sequence, then be able to
see what those movements translate into RCX code.
As it is you have to "reverse engineer" the movments in your head as to what
you want your robot to do. Then attempt to create the Code "foward engineer"
as to what you want the robot to do.
I say, run your robot through its paces realtime by hand, let the TMMC break
down the movements to Code, review the code to make sure slight differnces
intime are rounded up/off or balanced, then be able to send that polished
code to your RCX and then you cut the time it takes to write code by setting
an example or template for yourself based on what you want your robot to do.
> Instead of improving the hardware, a case could be made to improve the
> language support of RIS. But, with Robolab and other 3rd party solutions
> readily available (like NQC, BrickOS, pbForth, leJOS, etc...), the product
> may have reached the point of diminishing returns as concerns further R&D.
> Hopefully, we can continue to see revisions to the RIS software & SDKs that
> will support any new operating systems that M$ presumes to afflict us
> with...
Exactly. When I envision my Robot picking up an object then setting it back
down. I have to run tons of tests, making sure that the code replicates what
I intend the robot to do. If I could start out with a manual control, moving
my robot, then the TMCC watches my commands, generates code for what I just did.
Then I can go and tweak the code or revise it.
That would save time, and allow for a fast track to programming based on the
fact that you have just outlined what you want the robot to do.
Imagine, rolling a car around on the floor. Then knowing that when you place
the car down on the floor, it will repeat the same amount of movments. But,
allowing you the ablitity to view and edit the code that creates the movement.
Think of it like, ok, I want my robot to this, this, that and this. Then the
TMCC generates code for the movements you just "taught" the robot to do.
Am I on to something here?
> What could really be bad news is if TLC (knowingly or otherwise) adopts a
> "been there, done that" attitude and stops supporting the product because
> all enthusiam for it has died out (in the company). So many of TLC's
> products have relatively short shelf lives when compared to some other toys,
> I just hope TLC can continue to resist the temptation to do the same with
> RIS.
I hope not, but as I have said before, the tech in these products are the
essence of the future. They would be moving backwards if they did.
One of the worst things someone could say is that, "oh they used to make
cool stuff like that" That brings negative feelings. I hope they avoid that.
> I would certainly welcome RIS 3.0, but I'm not sure it's entirely necessary.
> Perhaps for marketing purposes, a repackaged and slightly enhanced RIS 3.0
> may be a possibility.
>
> Mark
I'd like to see an intelligent control center that helps add a bit on manual
prototyping as a human/RCX code generator of somesort More Motor Ports!
The Train folks I think would welcome the multiple motor contol center for
really cool Train Layouts....track switching and lighting.
e
http://members.aol.com/mylegomaster/
|
|
Message has 1 Reply:
 | | Re: RCX & RIS, a fading glory?
|
| Yes, you are on to something. Many industrial robots are "programmed" in much the same way. The operator "teaches" it which movements to make using manual controls, and then the robot can do those movements over and over again without error. In the (...) (22 years ago, 21-Aug-03, to lugnet.robotics)
|
Message is in Reply To:
| | Re: RCX & RIS, a fading glory?
|
| I see no reason why TLC couldn't just continue to manufacture RIS 2.0. Why continue with R&D to develop 3.0 if 2.0 is more than adequate to stimulate young minds into considering a career in engineering? 3 sensor inputs and 3 motor outputs are more (...) (22 years ago, 29-Jan-03, to lugnet.robotics)
|
33 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
