Threaded View
Subject
Joe Strout posted on August 4, 2007, 10:13 pm
Mark Miller has a neat article in the latest Servo about his experiments
with bipedal robots, which he builds from scratch. There are a lot of
design tidbits and insights -- some of them quite surprising, such as
the fact that he generally uses $3 stepper motors.
But while the article seems to imply that his robots can walk, he
doesn't actually come right out and say so, or discuss anything of the
algorithms they would be using to do so. No contact information or web
address was given, and Googling has turned up some news stories, but
nothing with much detail.
Does anybody know anything about these robots? Do they actually walk or
not, and how in the world could a 4-foot robot walk on $3 motors?
Thanks,
- Joe
--
"Polywell" fusion -- an approach to nuclear fusion that might actually work.
Learn more and discuss via: <http://www.strout.net/info/science/polywell/>
with bipedal robots, which he builds from scratch. There are a lot of
design tidbits and insights -- some of them quite surprising, such as
the fact that he generally uses $3 stepper motors.
But while the article seems to imply that his robots can walk, he
doesn't actually come right out and say so, or discuss anything of the
algorithms they would be using to do so. No contact information or web
address was given, and Googling has turned up some news stories, but
nothing with much detail.
Does anybody know anything about these robots? Do they actually walk or
not, and how in the world could a 4-foot robot walk on $3 motors?
Thanks,
- Joe
--
"Polywell" fusion -- an approach to nuclear fusion that might actually work.
Learn more and discuss via: <http://www.strout.net/info/science/polywell/>
John Nagle posted on
Re: Do Mark Miller's "androids" really walk?
Walking robots are easy, provided that they only have to walk on a
flat surface, move slowly, and have big feet. Wind-up toys can do that.
He does nice looking mechanical construction, though.
John Nagle
Re: Do Mark Miller's "androids" really walk?
To be fair, I haven't yet seen a wind-up toy that does that without
overlapping foot polygons.
Best,
- Joe
--
"Polywell" fusion -- an approach to nuclear fusion that might actually work.
Learn more and discuss via: <http://www.strout.net/info/science/polywell/>
castvee8 posted on
Re: Do Mark Miller's "androids" really walk?
My Androids do indeed walk-Mostly on flat level surfaces. I am going a
bit taller right now, almost 5'3 inches. I do in fact use cheap,
readily available stepper motors. Finding the right motor and gear
combination makes all the difference. I am writing an arm construction
article right now describing just such motors/gears all for a few
dollars each, which produces an arm that can lift a few lbs for about
50$.
The real challenge for me(building androids) was actually making the
joint system!
Mark
pogo posted on
Re: Do Mark Miller's "androids" really walk?
<http://www.strout.net/info/science/polywell/>
Cool. Do you have a web site so we can learn more ?
Thanks !
JCD
Re: Do Mark Miller's "androids" really walk?
No, I dont have a website-You can however e-mail me for any info.
Mark
Re: Do Mark Miller's "androids" really walk?
That's really cool. Do you have a web site, or any videos up?
I'm really looking forward to your arm article. (In fact, if you want
anybody to proofread it, I'd be happy to help.)
I'm just starting on a larger humanoid robot project -- though I'm
focussing on the upper torso for now, and not worrying about legs yet.
I'm still debating what to use for a frame, given a serious shortage of
time. I'm actually thinking about using Erector set parts, and just
trying to make sure they're covered enough not to look cheesy. :) I
might also consider Vex, though that'd mean investing in yet another
construction system, and my wife thinks I have too many already...
Best,
- Joe
--
"Polywell" fusion -- an approach to nuclear fusion that might actually work.
Learn more and discuss via: <http://www.strout.net/info/science/polywell/>
Re: Do Mark Miller's "androids" really walk?
<http://www.strout.net/info/science/polywell/>
How *well* do they walk ? Do they just slide their feet forward ? Is it
supported only by the two feet or is there a stand to stabilize it ?
Personally, I would like to see a follow-up to this article showing *how*
they walk. From the article, it *appears* that all you have done is
duplicate the joints and the general exterior shape of human legs. A side
view of the legs shown in Figure 2 showing the gears and mechanical linkages
would be an improvement - thus giving us a better idea of how you did it.
I'm just saying that if you could demonstrate more on how to get an android
like yours walking it would be a stronger inspiration.
I look forward to the article on the arm!
Just giving my 1.2 cents.
JCD
Re: Do Mark Miller's "androids" really walk?
Hopefully, I will get to all that and more in future articles.
I certainly didnt go to all that trouble to slide the feet along!
Site Timeline
- » inexpensive wheeled base for otherwise humanoid bot?
- — Next thread in » General Robotics Forum
-
- » Is Vex Receiver compatible with other RC gear ?
- — Previous thread in » General Robotics Forum
-
- » evoMUSART 2013: First CFP (with correct dates)
- — Newest thread in » General Robotics Forum
-
- » Bending 1/8 inch thick 7075-T6 aluminum
- — The site's Newest Thread. Posted in » General Metalworking
-
- » Rysunek techniczny-pytania
- — The site's Last Updated Thread. Posted in » Engineering Science (Polish)
-
XML Sitemap