Subject:
|
thoughts on dragsters
|
Newsgroups:
|
lugnet.org.ca.rtltoronto
|
Date:
|
Wed, 7 Mar 2001 04:11:18 GMT
|
Viewed:
|
526 times
|
| |
|
|
Ok. Let me start by admitting that I haven't been to the page listing
the proposed events yet, so maybe this is a pointless suggestion but
here goes anyway:
A few posts ago (I couldn't find it again) someone commented that
there wasn't much of an RCX related challenge in building a dragster.
That is to say, I could just as easily be done without an RCX.
Here's a thought that includes the cherished wall-as-a-finish-line
idea:
*Start* at the wall, race out to a mid-point line, and reverse back to
the wall again. That would require at least some programming in order
to detect the turn-around point and reverse the vehicle. Reversing the
vehicle will also require some skill given that vehicles with brakes
suddenly applied, or wheels thrown in full reverse tend to skid and
those skids tend not to be in straight lines. Braking too hard will
likely rotate the vehicle and send it off track. Walking the fine line
between reversing as quickly as possible while not losing traction is
probably going to be harder to do than one would think, given the
floor surface of the library probably isn't too simliar to what most
people have at home to practice with.
One could modify the race by declaring a winner to be the person who's
robot completed the circuit in the shortest time and award bonus
points to those who *DON'T* hit the wall on the way back (say 1000
points/distance to the wall in mm). Now *this* would require some
programming skill. How do you get that close to a wall without hitting
it? Use a light sensor? If you're moving at any reasonable speed you
won't have time to stop by the time the sensor reads the wall. The
second problem is that you need the light sensor to point at the floor
in order to detect the mid-point line. You also want the sensor at the
front of the vehicle at both the mid-point and at the wall, but
remember, we're reversing the vehicle, not doing a 180deg turn.
Decisions, decisions... You'll have to count the distance to the
mid-point with a rotation sensor and hope you can rely on the count on
the way back (definately better not skid when reversing). That should
tell you when you're close to the wall but if the big points only come
in the last few mm, you'll need something more accurate. The light
sensor would be good, but that's pointing at the floor and likely at
the other end of the dragster. Should you move the sensor within the
vehicle on the way back? That would mean extra weight and complexity.
I think there's definately a few challenges in this seemingly simple
idea.
Matthias Jetleb
VA3-MWJ
P.S. It occured to me as I wrote this: I haven't used the rotation
sensor yet - I'm assuming that it can be used at the same time as the
light sensor. I'm also assuming that only input 2 can be used for the
light sensor and that 2 light sensors cannot be wired in parallel.
|
|
Message has 1 Reply:
 | | Re: thoughts on dragsters
|
| (...) HI: I'm a little confused by what you asked here. all sensors can be used at the "same time" as every other sensor. the RCX can run up to 10 separate tasks. it would not be hard for the brick to have task 1 watch a light sensors on input 1, (...) (24 years ago, 7-Mar-01, to lugnet.org.ca.rtltoronto)
|
3 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
