Subject:
|
Re: Mindstorms NXT programming languages
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Fri, 13 Jan 2006 21:42:55 GMT
|
Viewed:
|
2567 times
|
| |
|
|
In lugnet.robotics, dan miller <danbmil99@yahoo.com> wrote:
> Point is, robotics research has gone way beyond line following with a PID.
> Roboticists have dealt with underpowered brains for years. Techniques have
> been worked out to work around that problem, by linking the robot to a more
> powerful brain. That brain is your computer. The link is bluetooth. The
> point of Python, Java, C++, Scheme (http://pyrorobotics.org/?page=Gyro), or
> any other high-level language (something other than C, which is basically
> glorified assembler) is to facilitate the design of complex, flexible
> algorithms that can enable more advanced behavior. What I would like to do
> with Python, or even with some new language (as I explained in a previous
> post), is to get as much autonomous functionality into the NXT as possible,
> and make it easy to do the rest on a remote machine. Imagine the robot
> having a dozen possible behavior patterns, which could be swapped in quickly
> from your laptop, depending on what the robot encounters. It's just an
> application of the trendy field of distributed computing.
>
> There's plenty to learn fiddling with lego pieces and RIS (or Labview on
> NXT), but if someone wants to move beyond that into some of the real
> cutting-edge stuff, it would be nice if there was a path to lead them there.
>
> -dan
You are right when you say the link is Bluetooth. With the RCX the link was IR
signals. As long as there is some communication protocol (the more standard,
the better) extremely complex logic can get into a robot's behavior.
With the RCX I was able to create a chess playing robot. I did it using the
standard firmware and NQC. Was all the computing on the PC? yes. Did it
reduce the RCX to a interface for the PC to my sensors/motors? yes. Does that
make my robot less autonomous? I don't think so. I have to program the
behavior of the robot, but I really doesn't matter where the brain and memory
are. (OK, it does a bit because of the time lost in communication, but for many
applications this is negligible)
Bluetooth will be far better than IR signals in every way. I would guess that
with Bluetooth we will be able to expose the NXT to just about any language you
want to program with. (along with a fast computer with lots o memory, etc.)
As for those talking about how much memory the NXT will have, a lot can be done
with the memory upgrade the NXT is getting (vs. RCX) And for what it can't do,
as long as there's enough computing power on the NXT to expose it to your PC,
who cares?
|
|
Message is in Reply To:
 | | Re: Mindstorms NXT programming languages
|
| What would I like to implement that won't fit in 256K... check out this list of modules from Pyro: # PyroModuleDirectControl # PyroModuleSequencingControl # PyroModuleBehaviorBasedControl # PyroModuleReinforcementLearning # PyroModuleNeuralNetworks (...) (19 years ago, 13-Jan-06, to lugnet.robotics)
|
52 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
