Subject:
|
Re: RotationNavigator accuracy
|
Newsgroups:
|
lugnet.robotics.rcx.java
|
Date:
|
Thu, 15 Jan 2004 09:36:13 GMT
|
Viewed:
|
5540 times
|
| |
|
|
Hi Robert,
thanks for your response. Indeed, the main problem seems to be slippage.
The more speed the motors produce, the bigger the slippage is and the
less is the resulting angle. Probably a tank is bad for odometric
navigation.
It is a pitty, that the articles are in Czech, I am still looking for a
good introduction to odometry. I do not know yet exactly
how the angle is calculatet in turns.
Regards
Michael
robert seifert wrote:
> > How can this be, if it calculates with rotation sensor values,
> > wich do not depend on motor power?
> >
> >
>
> rotation sensor counts tread rotation, but, because of skidding, the tank
> doesn't turn exactly the same way as treads do
>
>
>
> > RotationNavigator nav=new RotationNavigator(2.64f, 9.5f, 1.0f);
> >
> > The drive length is the length from one middle of the tread to the
> > middle of another.
> >
> >
> >
> rotation point position/drive length depends on tread width, center of weight
> position, speed and geometry of treads and on friction coefficient, so your
> value aproximates it very roughly. try to estimate it better by experimenting -
> just let your tank perform series of turns and watch it.. then you should be
> able to pinpoint and measure position of center of it's rotation.
>
>
>
> > Especially I do not understand, why the turning accuracy depends on
> > motor power and why the resulting turning angle is so different
> > depending on motor power.
> >
> >
>
> i gues that's because of different slipping, but i am not sure
>
> hope this helps
>
> robert
>
>
> > Regards
> > Michael
> >
> >
|
|
Message has 1 Reply:
 | | Re: RotationNavigator accuracy
|
| "Michael Wisse" <mwisse@hs-harz.de> wrote in message news:40065F0D.70202@...harz.de... (...) a (...) Check out (URL) the "Peeves" Mindstorms robot. If page is not accessible then just do a search on "peeves mindstorm lego" and you'll probably find (...) (21 years ago, 15-Jan-04, to lugnet.robotics.rcx.java)
|
Message is in Reply To:
| | Re: RotationNavigator accuracy
|
| (...) rotation sensor counts tread rotation, but, because of skidding, the tank doesn't turn exactly the same way as treads do (...) rotation point position/drive length depends on tread width, center of weight position, speed and geometry of treads (...) (21 years ago, 14-Jan-04, to lugnet.robotics.rcx.java)
|
5 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
|
|
|
|
