Page 1 of 3 123 LastLast
Results 1 to 10 of 26

Thread: Biped ankles

  1. #1
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Biped ankles

    Lately I've been thinking that the common practice of using the same servos in a biped's ankles as the rest of the robot might not be quite optimal.

    Most of the DOF in a biped use a relatively large travel range of motion. That's not true for the ankle DOFs. I'm not a very flexible guy, but I seem to be able to get around quite well with only about 40 degrees of ankle pitch and even less ankle roll. In giving up range, the servo is giving up precision and because the travel range is so small you could probably get away with less servo speed.

    Because of this I am thinking that it might make sense to use some sort of worm drive setup for ankles actuation. It seems like you could make a much more compact and more lightweight actuator using a worm drive setup. The disadvantages I can think of are relatively poor worm drive efficiency and zero compliance. I'm thinking that neither might end up being a problem in practice. Ankles really don't usually make large movements so efficiency might not matter as much there. For compliance I think it also might be fine because the ankle is not isolated and compliance can easily come from servos up the chain ( such as the knee ) as well as sensors ( in the non-mechanical sense ) located on the foot itself.

    I'm wondering what other people's thoughts on this are.

  2. #2
    Join Date
    Dec 2007
    Location
    Whidbey Island, WA
    Posts
    1,718
    Images
    456
    Rep Power
    104

    Re: Biped ankles

    I could see using some sort of lever system, that way you could move the weight of the ankle actuators to some other part of the robot, increase the strength of the actuator (torque) and not sacrifice too much in the way of speed. This might have the added benifit of reducing the strain on the hip servos. Another question is are the benefits of this sort of system of advantage in robots below a certain scale? Probably limited at robots zog sized or smaller...not really sure where to draw the line though. As for worm drive, might it be too slow? It would have the benefit of not drifting under load... I don't know...just thinking aloud.

    DB

  3. #3
    zoomkat Guest

    Re: Biped ankles

    With a linkage from the servo to the outside of the foot, the full 180 of the servo could be translated to smaller degrees of movement in the ankle. Something similar at below link.

    http://davidbuckley.net/DB/Inspire.htm

  4. #4
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Re: Biped ankles

    I'm just thinking aloud too. haha.

    I've seen a few bots relocating actuators into the lower leg. It makes sense considering animal/human legs do the same thing. The ones I've seen use pretty lightweight linkages, but maybe that's a good idea which increases compliance. I guess it would be perfectly reasonable to build more substantial linkages too.

    Why do you think worm drive would be limited to smaller bots? I was actually thinking about this for a larger bot. I don't think a worm drive would necessarily be to slow. I was actually thinking it would make sense to gear a motor down with a few spur gears with the output ending in a lower pitch worm gear.

  5. #5
    Join Date
    Dec 2007
    Location
    Whidbey Island, WA
    Posts
    1,718
    Images
    456
    Rep Power
    104

    Re: Biped ankles

    No no...sorry...I meant that worm drive would be better on larger bots not smaller ones because on smaller ones the gain of strength would be outweighed by the gain in weight...unless the worm drive assembly was really light.

    DB

  6. #6
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Re: Biped ankles

    zoomkat. I'm not sure it would be possible to come up with a more appropriate link. bravo!

  7. #7
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Re: Biped ankles

    I'm thinking the worm drive would be smaller/lighter for smaller and larger robots because I think you can get away with less torque overall. I'm thinking that the ankle servos might actually need a lot less dynamic torque and what is instead important is holding torque. My thinking is this.. Imagine a very heavy pole. It's quite easy to knock it over, even from near the bottom. However it's not so easy to stop or hold it once it is tipped. Seems like this is an analogous to a ankle/leg setup. Holding torque on a worm drive is insane so it might actually be the better actuator choice for this reason too.

  8. Re: Biped ankles

    Im building this ankle design at the moment. The plan is that this design is better because both the motors are used to do all the work. ie Both are used when tilting forward, backwards, left right etc. The ankle is a lot smaller than drawn here.


  9. #9
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Re: Biped ankles

    That looks really great. It looks like it has a lot of potential. Your insight about both motors sharing the load is very interesting. That had not occurred to me.

    I see you're also looking into a toe. Something else that needs its own thread IMO!
    Last edited by billyzelsnack; 04-28-2010 at 10:29 AM.

  10. #10
    Join Date
    Aug 2008
    Location
    Iowa City, IA
    Posts
    390
    Rep Power
    53

    Re: Biped ankles

    On the subject of worm-drive.. I ran across an issue that was immediately obviously after receiving a surplus gearbox in the mail.. Backlash.. Lots of it. The particular gearbox was gear down with spur gears and used a final larger pitch worm drive to rotate the output shaft 90 degrees. In theory less gearing down and a finer pitch worm gear could be used, but I'm not just getting a good vibe. Yeah.. So scientific there. haha.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Biped Video Challenge....
    By Atom in forum Humanoids, Walkers & Crawlers
    Replies: 34
    Last Post: 04-18-2011, 02:44 PM
  2. For Sale: Biped Scout with 12x NEW 5990TG
    By tom_chang79 in forum Garage Sale
    Replies: 10
    Last Post: 04-16-2011, 02:13 AM
  3. How to scale up to a larger biped
    By billyzelsnack in forum Humanoids, Walkers & Crawlers
    Replies: 75
    Last Post: 04-13-2010, 07:25 AM
  4. Project help!.kinematics analysis for my biped robot
    By kishy22 in forum Mechanics / Construction
    Replies: 5
    Last Post: 01-25-2009, 01:27 AM
  5. New 6DOF SES Biped Leg
    By LinuxGuy in forum Humanoids, Walkers & Crawlers
    Replies: 14
    Last Post: 04-26-2008, 06:18 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •