Subject:
|
RE: Rack-and-pinion steering line tracker
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Sat, 8 Jan 2000 01:19:33 GMT
|
Original-From:
|
Tilman Sporkert <tilman@!ihatespam!activesw.com>
|
Viewed:
|
502 times
|
| |
|
|
I too spent quite some time trying to build a steering mechanism from standard
Mindstorms parts. After many rebuilds I still wasn't quite satisfied. The final
version actually worked pretty well, but was much too bulky (not as bulky as
Doug's version however :-) ), and the wheels were tilted inward too much for my
taste. So I decided to see what I could do with some more specialized Lego
parts. I could have just bought the service pack with the various Lego steering
mechanism parts from S@H. But no, I decided to use it as an excuse to buy all
sorts of Technic sets. Along the way I learned that Lego can't make up their
mind about how to build a steering mechanism either. There are so many
variations!
About the return to center problem: My solution is to incorporate the light
sensor into the steering mechanism. Mount it at the end of a short arm pointing
forward, ahead of your wheels. This has two advantages:
- You will get a much smoother steering vehicle, with gradual steering
movements.
- You don't have to worry about centering the steering mechanism.
With the traditional approach, you will be sensing that you are off the track.
But you don't really know how far to turn the wheels to stay on the track. So
you make a best guess, turn too far, until you are back on center. Now you have
to figure out how to go straight again.
With my approach, you keep the light sensor on the track all the time, and the
vehicle sort of follows it. As soon as the sensor moves off center, you start
steering to keep the sensor on the track. The steering system is now part of a
direct feedback loop, and will always point in the right direction.
This of course doesn't make the steering system any easier. You now have to turn
three things - the two wheels, and the arm that holds the sensor.
Tilman
> I spent a _lot_ of time trying to get a non-sensored R&P steering system to
> work, using pulleys, rubber bands (both as drives and springs), and
> you-name-it! I finally gave up, but did manage to build one with only _one_
> sensor that was very effective.
>
> See http://www.wordsmithdigital.com/mindstorms/ for a scan and my described
> history of running my head into the wall. Should get you started in the
> right direction--or avoid getting you started in the wrong direction.
>
> Ironically, I've started playing with roboticizing the 8479 as well!
>
> --Doug Wilcox
|
|
Message has 1 Reply:
 | | Re: Rack-and-pinion steering line tracker
|
| I'm working on my first original robot, having built the 3 in the Constructopedia and modified them a bit. I'm currently on my 4th R&P design: the first two just didn't work, and the 3rd had a turning circle larger than my living room. It (...) (24 years ago, 9-Jan-00, to lugnet.robotics)
|
Message is in Reply To:
| | Re: Rack-and-pinion steering line tracker
|
| I spent a _lot_ of time trying to get a non-sensored R&P steering system to work, using pulleys, rubber bands (both as drives and springs), and you-name-it! I finally gave up, but did manage to build one with only _one_ sensor that was very (...) (24 years ago, 7-Jan-00, to lugnet.robotics)
|
3 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
|
|
|
|
