Subject:
|
Re: Using the Differential to drive straight.
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 15 Jan 2002 01:42:11 GMT
|
Viewed:
|
640 times
|
| |
|
|
Mario wrote:
> Actuially the solution that we describe in our book is similar but not
> identical to Dave's one. Starting from the well known setup in which a
> differential gear is used to measure the difference in speed between the
> main wheels of a differential drive - so that it doesn't rotate when the
> vehicle goes perfectly straight - we suggest that it's possible to place a
> motor connected to the body of that diff.gear so as to *prevent* it from
> turning. In our solution this third motor has to stay OFF (electrically
> braked) during straight motion, and it has to FLOAT during turns.
> Though this solution works, I would use a mechanical lock if you want to be
> sure that your robot travels perfectly straight, for example to attend some
> kind of competition about positioning.
My implementation did something similar but used the 3rd motor in the
OFF
(ie. high resitance to turning) setting during turns to make sure that
both the A and B side wheels turned the same amount. This helps to make
sure that the skid/steer turns are on a dime as it were.
When the 3rd motor is set to FWD or REV, a bit of added torque is sent
to the differential is added to whatever torque being generated by both
A and B motors. There is no guarantee of the A and B sides being locked
together as in Mario's implementation (where one of the Motor/drive
shaft
inputs is reversed before entering the differential) which becomes in
effect a selectable axle lock. But this added torque seems to help in
evening out torque differences between the A and B motors (or maybe this
is just a conincidence and my motors are pretty closely matched to start
with? Maybe a quirk of my particular drivetrain arrangement?).
I'll do a few more test runs with straight line travel with the C motor
on FLOAT and with it running and see what it looks like. But the
travelling in a square test I described in a previous posts does look
lots better with the C motor OFF/locked.
Dave
|
|
Message is in Reply To:
 | | Re: Using the Differential to drive straight.
|
| (...) Actuially the solution that we describe in our book is similar but not identical to Dave's one. Starting from the well known setup in which a differential gear is used to measure the difference in speed between the main wheels of a (...) (22 years ago, 14-Jan-02, to lugnet.robotics)
|
10 Messages in This Thread:
 
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
