Group Actions
Engagement
  • Author Author: genebren
  • Date Created: Date Created
  • Views 1834 views
  • Likes 14 likes
  • Comments 8 comments
Related
Recommended

Walky the Biped Robot - WalkyII gets his legs

genebren
image

Movers and Shakers

Enter Your Electronics & Design Project for Your Chance to Win a $100 Shopping Cart!

Back to The Project14 homepage image

Project14 Home
Monthly Themes
Monthly Theme Poll

 

Previous post in the sequence:

Walky the Biped Robot - The prequel

Walky the Biped Robot - Power pack

Walky the Biped Robot - Oops, Walky has an accident!

Walky the Biped Robot - A new hope (actually a new body and battery)

Walky the Biped Robot - WalkyII gets some eyes.

 

Today the 3D printathon has ended.  I finally have some peace and quiet at my desk.  At 2:00 PM, the final piece of WalkyII was printed (final for now?).  After a bit of assembly, WalkyII can now stand on his own.  Things went pretty well.  I needed to clean-up a few holes, tighten this, loosen that...... But in the end, it looks exactly like what I planned.

imageimageimage

Once I assembled him, I decided to see how well he could walk.  I powered the 'original' controller board from the new power pack and hit the switch.  WalkyII did not do too well.  He toddled around like a drunken sailor and would tip over after each step (if not for my helping hand).  I appears that the center of gravity has shifted a bit from the original Walky.  Also some of the joints were a little loose, mostly the 'ankles'.  I did a little rework on the 'ankles' and adjusted the servo linkages a bit to shift feet a little further back.  I also adjusted the front linkages to get the feet a little flatter to the table (the above pictures are post reworks).  I still need to do something to move the feet a little further apart, maybe a slider positioned between the legs and the body, along the bottom edge, to spread the legs out a bit.

 

I did capture a few short clips of WalkyII walking.  While he is not exactly graceful, he can now toddle about without falling over.  I will need to fine tune his gait some, as he has a tendency to walk in circles.  This is all stuff that I can address when I start coding his new controller board.

 

Here are some of the 3D views of his legs, feet and ankles:

imageimageimageimage

All of the pivot points on the legs are done using pressed in nylon shoulder spacer, in an attempt to have a free movement with a minimal amount of wobble.  The majority of the hardware to bolt everything together is 2-56 screws.  On the pivoting leg joints I used nylock nuts to lock the nuts on, without over tightening the joint.

 

So far, so good.  But I do see a lot of work coming up writing the firmware for the controller and finishing up the firmware on the power pack.  (not to mention that I am currently juggling three consulting gigs).

 

Thanks for reading along.  More to come.

Gene

Parents

  • John and Mark,

     

    I actually did put rubber dots on the original version of the robot to stabilize the forward motion.  It is a delicate balance between having the feet too slick or too sticky.  The normal forward movement sequence is to tilt first, then drive the legs (the lifted one forward and the weight bearing back).  But to turn the robot you skip the tilt and drive the legs, then tilt and drive the legs.  If the feet and too sticky, the robot does not pivot correctly and it does not turn.  I purchase a variety of pads (vinyl, foam and cork) that I plan to test, trying to get the proper compromise. The new feet are slightly more textured (printed) versus the old feet which were very slick.

     

    One of the biggest issues right now have to do with the proper centering of the servos.  When I disassembled the earlier version, removed the servo horns and did not mark the servos.  So when I re-assembled, the centers were not correct and some of the actions did not balance well.  Also the center of gravity has shifted and I still don't have it centered correctly yet.  I have a lot of tuning to do.  The new controller firmware will help, as I will be able to tune the servo centers, and the amount of tilt and leg movement to match all of the other changes in the robot's construction.

     

    Thanks,

    Gene

  • in reply to genebren

    Hi Gene,

    What you describe is a principal that I have come to accept as a factor in all repairs and modifications that I make to devices. Always expect and try to anticipate the side effects of any change. Many times I have made a change to a design in order to correct an inconvenience of a failure only to find that I created another problem. Sometimes it is like dominoes falling over. Good luck and enjoy the process of perfecting Walky's gait.

     

    What about suspending a weight from his bottom center or putting weights on his feet to lower the center of gravity?

     

    John

Comment
  • in reply to genebren

    Hi Gene,

    What you describe is a principal that I have come to accept as a factor in all repairs and modifications that I make to devices. Always expect and try to anticipate the side effects of any change. Many times I have made a change to a design in order to correct an inconvenience of a failure only to find that I created another problem. Sometimes it is like dominoes falling over. Good luck and enjoy the process of perfecting Walky's gait.

     

    What about suspending a weight from his bottom center or putting weights on his feet to lower the center of gravity?

     

    John

Children
No Data