Subject:
|
Re: rtlToronto15 January issues
|
Newsgroups:
|
lugnet.org.ca.rtltoronto
|
Date:
|
Tue, 26 Aug 2003 03:00:17 GMT
|
Viewed:
|
1032 times
|
| |
|
|
Calum Tsang wrote:
> Back on topic, The major problem I see with rtl15 is maze strength. If we have
> terminated bad branches, I'm not sure the thing will stay together. Hence, I
> wanted to build it stronger, using the bad branches as part of the structure.
>
> We can still do bad branches as terminations, but use them as part of the
> structure, just mark off the bad paths.
>
> Or we can be nice and offer secondary paths on the bad branches.
>
>
I think you've got a good idea here. If we build it right, we wouldn't
have to rebuild it to have a different maze for subsequent rounds, we
could just mark different paths through the maze.
I don't like the idea of having more then one correct path. I think if
you go down the wrong path it should be considered a failure. Scored up
to the point of the wrong turn.
We could consider making flanges that could be clamped into the middle
of the wrong paths. This would give a dead end the robot could detect.
The robot could then go back and try again. A second chance of sorts if
you miss read the markings.
Derek
|
|
Message is in Reply To:
 | | Re: rtlToronto15 January issues
|
| (...) Sad thing is I learned CPM techniques in first year, but the best implementation I've ever seen is in our GPS lab, where they put on the whiteboard whoever's behind the 8 ball. It gets people working FAST, and no need for CPM charting either. (...) (21 years ago, 25-Aug-03, to lugnet.org.ca.rtltoronto)
|
30 Messages in This Thread:
 
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
