Subject:
|
Re: Classroom experiments gone awry
|
Newsgroups:
|
lugnet.robotics.nxt
|
Date:
|
Sun, 29 Apr 2007 02:02:08 GMT
|
Viewed:
|
19586 times
|
| |
|
|
In lugnet.robotics.nxt, <dickswan@sbcglobal.net> wrote:
> Your results are not at all what I intuitively expect.
Steering under NXT-G is... not exactly intuitive. I tried some experiments a
whle back, but I never even tried to play with the steering slider because I had
no patience for rough approximations (like trying to see if I had moved the
slider exactly one "click"). So I immediately went on to wiring in values to the
steering port, and charting the result. Conclusions?
The further the steering slider moves from zero, the less the inner wheel
turns, reaching zero travel at not 50, but at about 35. For steering values
larger than this (positive or negative), the inner wheel travels backward. The
travel of this inner wheel reaches a maximum when the steering value is not 100,
but 64. Throughout the whole range, the outer wheel total travel distance does
not change. The good news is the inner wheel travel distance is fairly linear
between 0 and +/-64.
The easist way to see this is try it: write a program that wires in a
specific value, and then look at the number of degrees each wheel has turned
(the feedback pane in the configuration panel is ideal for this). At very rough
resolution, for a duration of 1000 degrees...
Steering B C
100 -1000 1000 |
90 -1000 1000 | turn
80 -1000 1000 | in
70 -1000 1000 | place
64 -1000 1000 |
60 -775 1000 |
50 -460 1000 | tight turn
40 -170 1000 |
35 0 1000 Turn around one wheel
30 145 1000 |
20 420 1000 | wide turn
10 730 1000 |
0 1000 1000 No turn at all
-10 1000 730
-20 1000 420
-30 1000 145
--etc--
--
Brian Davis
|
|
Message has 1 Reply:
 | | Re: Classroom experiments gone awry
|
| (...) I placed a dollar bill with one corner of it pinned under a chair leg. The challenge was to start from the spot and circle around the outside of the legs. Whichever team can do it in less than two tries claims the dollar. I think my dollar is (...) (18 years ago, 29-Apr-07, to lugnet.robotics.nxt)
|
Message is in Reply To:
| | RE: Classroom experiments gone awry
|
| Your results are not at all what I intuitively expect. From your previous posts, I think the "turn ratio" settings you've used range from 0 to +100 in steps of 10. The following table is what I expected. Turn Inner Expected Observed Ratio Wheel (...) (18 years ago, 28-Apr-07, to lugnet.robotics.nxt)
|
9 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
|
|
|
|
