Subject:
|
Re: Building Competition LEGO Robots
|
Newsgroups:
|
lugnet.org.ca.rtltoronto
|
Date:
|
Thu, 15 May 2003 16:08:56 GMT
|
Viewed:
|
525 times
|
| |
|
|
> Remember: A robot is complete not when there is nothing left to add,
> but rather when there is nothing left to take away.
>
> Jeff E
This is a good quote. It is open to interpretation though. For example,
cross bracing could easily be taken away without altering the robots
functionality, but I think it is essential. I think there is a balance
between holding the robot together and making it bomb proof. I tend to errr
on the side of bomb-proofing, and I think Dave does as well. But, making
something overly strong does add to its reliability. Our motto would be
"cross-brace, then cross-brace some more"
Anyway, I would agree with the simplicity rule, but I am not good at
following it. Case in point: My block stacking mechanism. The basic elevator
mechanism had three competitions worth of refinement and testing. The first
two versions used a complex spring tension system and ratcheting wheels to
grab the blocks. The final version was much simpler and worked by gravity
and a cam action between the wheels and the blocks. So it was simpler, but
not nearly simple enough. At the end of the day, it was just too finicky and
the blocks had to be in perfect alignment before being lifted for it to
work. On my "bench tests" it worked fine. On the back of Steves robot, it
barfed. This brings me to another point: testing.
Testing robots is absolutely essential, probably the most important thing
one can do. And unlike my stacker, testing has to be done in an environment
that is as realistic as possible, if not completely replicating the
competition environment altogether. This helped enormously with the monkey
bars game. I built a full size ladder even though a half size ladder would
have been fine. The robot fell many times before it actually started
working. When It did, my dad went a step further and built a ladder that was
crappy on purpose. He wanted to see if the mechanism could handle offset
rungs and inaccurate spacings. So when time permits, extend your testing to
the worst case scenario.
Speaking of all this reminds me that it has been a while since I put
together a real robot for a competition. I recently built the box for the
rice game and have tried out a few ideas. All of my initial concepts proved
to be overly complex. I threw together some quick mock ups and realized that
there were too many different mechanisms that were interdependant so if one
of them failed, the whole robot would fail. Is anyone else finding this game
to be quite challenging?
Rob
>
>
> Steve Hassenplug wrote:
> >
> > Several of you already know, I'm writing a book on Building Competition • LEGO
> > Robots.
> >
> > I plan on including some things that _I_ consider important when • building
> > LEGO robots, but it suddenly hit me... What does everyone else think is
> > important?
> >
> > I want to put up a poll, to rank what people from LUGNET think is • important,
> > but I need to figure out what to list on the poll. I have some ideas, • but
> > I'd like to find out what rtlToronto members think.
> >
> > SO...
> >
> > What do YOU think is important when building a [competition] LEGO robot?
> >
> > Please take a second (before reading other people's ideas) and think of • an
> > answer.
> >
> > Feel free to post here, or e-mail me directly.
> >
> > Thanks
> > Steve
> > Hassenplug@mail.com
|
|
Message is in Reply To:
 | | Re: Building Competition LEGO Robots
|
| Easy - the most important consideration is to keep it simple. Every feature or capability or software response you build in lets your bot DO something, but it adds failure modes. So does adding the feature improve your chance of winning more than (...) (22 years ago, 14-May-03, to lugnet.org.ca.rtltoronto)
|
11 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
