Page 1 of 2 12 LastLast
Results 1 to 10 of 11

Thread: Novel leg/chassis design

  1. Novel leg/chassis design

    Hi everyone

    I saw the Tested video from the Bay Area Maker Fare and got interested. The bots look awesome, but $480CADplustax & shipping for just the leg servos alone is a bit too much to drop on something I've never done before. Especially if I want to try against another person I'll probably have to make two.

    It sounds like the main reason to use the dynamixel is their ability to support the weight of the robot. Has anyone tried legs that support the weight of the robot mechanically? The concept shown below uses two steppers and a five bar linkage to position the foot horizontally. The foot is to be actuated through the square holes in the leg, I haven't worked that part out yet. The goal is to support the weight of the robot without putting continual load on the motors.

    comments and criticisms welcome, I'm very much making this up as I go.



  2. #2

    Re: Novel leg/chassis design

    I'm not seeing a picture?

    Some contestants have used mechanical advantage. There was a nice two-legged robot a while back that used four servos total (two per leg, for "extent" and "front/back" articulation,) and a slip ring, for a very high vantage point, and 360 degree field of articulation.

    Note that most four-legged robots will want three degrees per leg, which gives you front/back, up/down, and left/right articulation, which allows you to turn without slipping, to strafe, and so on. If you can build a linkage that goes front/back without the arc of just swinging around the hip, or if you can live with the slipping, you may be able to get away with two degrees of freedom per leg. Especially if you also rest on the belly while "scooting" you could probably make that work just fine.

    Another option is to still use articulated legs, but add torsion springs or some other form of stored energy/elastics, so the robot stands neutrally without power, and only uses the servos for articulation, not load. Even for robots with servos that can carry the load, springs help reduce power consumption and increase runtime on a single battery charge!

  3. Re: Novel leg/chassis design

    Hi Jwatte

    Thanks for the feedback, It is good to hear that I'm not off on some wild goose chase.

    I'm still getting used to how the forum works, the pictures should be here now
    The highlighted circle is ~47mm across. I figure if all foot motions are contained within it then the motions you described should be possible.

    Cheers

    Click image for larger version. 

Name:	MechWarfare Mech5 GA Top.JPG 
Views:	27 
Size:	88.1 KB 
ID:	7577
    Click image for larger version. 

Name:	MechWarfare Mech5 GA Iso.jpg 
Views:	25 
Size:	34.5 KB 
ID:	7576

  4. #4
    Join Date
    Sep 2010
    Posts
    205
    Images
    16
    Rep Power
    37

    Re: Novel leg/chassis design

    David,

    I created a design somewhat similar in concept a long time ago, there is no reason it shouldn't work. I ended creating a 2dof per leg design that uses dynamixels in a more traditional configuration, but they are oriented such that there is minimal load on the servos in terms of continuous torque holding up the robot weight. Both of the immortal designs use it (the original had 4dof per leg but I changed it to 2dof), the only consideration you need to make is that the feet design needs to be well thought out because you want it to slip.

  5. #5

    Re: Novel leg/chassis design

    Given that linkage, you should be able to move the feet in X/Y, but you also need something that changes which foot has most pressure (i e, lifting a foot.)

  6. #6
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,282
    Images
    27
    Rep Power
    275

    Re: Novel leg/chassis design

    Quote Originally Posted by jwatte View Post
    Given that linkage, you should be able to move the feet in X/Y, but you also need something that changes which foot has most pressure (i e, lifting a foot.)
    Which is the undecided factor that goes in the square tube cutout in the actuated end of the linkages. My first thought for that was solenoid, then linear actuator (threaded rod coupled to a small motor with a mating nut in a foot/tube prevented from rotating with the threaded rod), but currently thinking small brushed dc motor in a system of pulleys would be interesting to see.

    The pulley system I'm thinking of could be two ways: foot on a linear rod normally pushed down by a spring (only activate motor to lift foot by loading/compressing the spring beyond the robot's weight on that foot) or foot on a linear rod normally pulled up by a spring (activate motor to extend the foot, but requires the motor to stall to keep the foot in ground contact and support the robot's weight plus spring load). A normally-on brake clamping down on the rod could minimize the continuous load on the motor when trying to maintain the non-default state of the foot (i.e. "default" being determined by the normal state of the spring when motor unpowered).

    Hmm. Yet another thing to maybe try when I have the time and energy.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  7. #7

    Re: Novel leg/chassis design

    I like the solenoid idea! You only need to activate it while the foot is "up" and presumably there would be a spring pushing it down with just enough force to keep it extended against half of the weight of the bot.
    But any original walking linkage is cool to see, so have at it!

  8. #8
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,282
    Images
    27
    Rep Power
    275

    Re: Novel leg/chassis design

    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  9. #9

    Re: Novel leg/chassis design

    And of course, my first-ever walking robot. A biped with only three servos!

    https://www.youtube.com/watch?v=SMXx6VZVjck

    (Actually, second-ever -- I built a smaller version in plywood with smaller servos first as proof-of-concept)

  10. #10
    Join Date
    May 2010
    Location
    Princeton University
    Posts
    349
    Rep Power
    36

    Re: Novel leg/chassis design

    There have been a couple viable three servo entries similar to the above. Mine for Shep Robo 2015 specifically did it for cost reasons. We managed with two AX and one MX, and here is a project that managed with just 3 AX servos.
    http://forums.trossenrobotics.com/sh...light=twitchmx

Thread Information

Users Browsing this Thread

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

Similar Threads

  1. Project Chassis
    By SteamAutomaton in forum Humanoids, Walkers & Crawlers
    Replies: 46
    Last Post: 11-29-2012, 12:08 AM
  2. Project Best chassis?
    By FieldDoc in forum Rovers
    Replies: 20
    Last Post: 08-13-2012, 10:46 AM
  3. FS: NEW Lynxmotion Tri-Track Chassis
    By charleyshf in forum Garage Sale
    Replies: 0
    Last Post: 03-11-2011, 09:22 AM
  4. Project Brainbot - Navigation Chassis
    By JonHylands in forum Project Showcase
    Replies: 18
    Last Post: 05-28-2010, 05:20 PM
  5. Looking to buy lynxmotion tri track chassis
    By Spartan001 in forum Garage Sale
    Replies: 5
    Last Post: 10-12-2009, 11:11 PM

Tags for this Thread

Posting Permissions

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