Subject:
|
Re: Walkers
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 13 Feb 2001 23:18:19 GMT
|
Original-From:
|
Josh Herman <maverick@infoconex#AvoidSpam#.com>
|
Viewed:
|
750 times
|
| |
|
|
I don't know much about mechanics or walkers, but what about a gyro device
that's more stable than a rotational sensor. Surely someone out there would
know how to use one of these to "balance" a walker. Of course this means
powering the gyro to spin constantly...
Josh
----- Original Message -----
From: Harley Myler <h.myler@myler.org>
To: <sjbaker1@airmail.net>
Cc: <lego-robotics@crynwr.com>
Sent: Saturday, February 10, 2001 10:03 AM
Subject: Re: Walkers
> At 5:37 PM -0600 2/9/01, Steve Baker wrote:
> > Harley Myler wrote:
> >
> > > What you really want to ask here is "Has anyone built a 2-legged • walker with balance?".
> > > I would say it is unlikely that one has been built, but certainly not • impossible.
> > > For a balancing mechanism you need position sensing, which could be • done with a
> > > rotary (distance) counter and a pendulum.
> >
> > I tried that in my attempts to build a two wheeled robot (as in a • bicycle)...but
> > the problem with rotation sensors (as has been discussed *extensively* • before)
> > is that it drops or gains counts sometimes when it's being rotated • slowly.
>
> My post was to clarify the walker question, the suggestion of using the • position sensor was just that.
>
> > That means that you really have gear up the pendulum's output in order to
> > spin the rotation sensor quickly - and that gearing slows the swing of • the
> > pendulum to the point of uselessness...unless it's VERY heavy.
> >
> > I had better luck with a two-coloured pendulum and a light sensor...but
> > there were many other problems - and I never did get it to work.
>
> The walker problem is not trivial, as you've discovered.
>
> > Incidentally, the way most bipeds walk is by leaning forwards so their
> > Center of gravity is ahead of their feet - and then sticking out a foot
> > to prevent themselves from falling. It follows that balancing the robot
> > may not be the best strategy here.
> >
> > --
> > Steve Baker HomeEmail: <sjbaker1@airmail.net>
>
> That is a correct observation, but no one wants a walker that can't stop • and smell the roses--for that you need balance.
>
> Harley Myler
> http://macmyler.engr.ucf.edu
>
|
|
Message has 2 Replies:
 | | Re: Walkers
|
| (...) The possibilities are endless. I would start with the sensor and move on from there. The stability and accuracy (precision) requirements will depend heavily on the robot configuration. It does not need to have pointed feet like a ballerina (...) (24 years ago, 14-Feb-01, to lugnet.robotics)
| | | Re: Walkers
|
| Has anyone considered adapting one of the tiny gyros used to stabalize remote control helicoptors ? These work by modifying the pulse stream that controls a standard R/C servo. - pete. (...) -- Pete Sevcik sevcik@flash.net Techno-stuff Robotics (...) (24 years ago, 15-Feb-01, to lugnet.robotics)
|
Message is in Reply To:
| | Re: Walkers
|
| (...) My post was to clarify the walker question, the suggestion of using the position sensor was just that. (...) The walker problem is not trivial, as you've discovered. (...) That is a correct observation, but no one wants a walker that can't (...) (24 years ago, 10-Feb-01, to lugnet.robotics)
|
40 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
|
|
|
|
