Ah yeah, I saw the Tar servo code, and figured it won’t be needed. From some readings, it seems that some “toe” servos might be better. Maybe later.
Your Wolf definitely gets around! Almost Chaplinesque! But by not using a yaw servo (which I have). The roll servo should come in handy at some time, but I don’t think it is being used now. Is that true? I’ll follow the calcs. So that leaves your three servos in the Z plane, and I’m left with two. I might have a problem.
there shouldn’t be a problem as Wolf will walk with out the tars. in theory! i only used them to return the foot flat to the floor.
as you have a smaller foot print than im using you would need to calculate the body sway to shift the cog. this is were i gave up and went with the flat foot.
if you wish to take this route, i can point you in the right direction. but its tricky.!
Oh that’s right. You do (did) have those LARGE feet! I just have the ends of the rubber-tipped SES tubes. Maybe I could add some temporary feet? I was thinking of shortening the tubes, or even removing them. But I want to add foot sensors.
So how exactly are you turning? Tank style? I don’t see how you can turn (yaw) the leg. I suppose it isn’t absolutely necessary. I’m split between too many projects!
I didn’t want to contend with body sway at this point.
properly best to watch the vid on how it turns.
i have to change gait for this.
yeah i was going to go for the foot sensors too. i am still working on the gait so i can re-add the tubes with rubber caps.
im finding that the backside of the “C” bracket is the right size for stability. i may make something of the same sorta size but incorporate a sensors of some sort.
Is the foot servo needed for anything besides squatting down? If not you might eliminate it to reduce the servo count by four, keeping the foot angle the same. You might also replace the flat foot pad with a horozontal piece of round tubing ~the same width as the foot pad. Should provide the same side tipping protection as the flat foot.
Good question, and some interesting thoughts.
The “Ankle” servo was originally designed to place the foot flat to the ground at all times. im no longer using the large foot but i have found its much more stable with the ankle servo still doing its job. i have already eliminated 4 servos by removing the Coxa’s so i have been looking at saving power and weight, but removing the Ankles wouldn’t be advised due to my findings.
the tube idea would work well but it wouldn’t look as nice.
your last video was really impressive. I am wondering if it might be more stable if you added some sort of rubbery grip on the feet. The movement was really incredible and for some reason it made me think of this video game called shadow of the colosus. Great work!
i have found that walking on tiles is best for this robot. strangely, adding rubber grip doesnt help at all. the feet grip too much not allowing it to turn or make smooth motions. i am using a felt back tape stuck to the bottom to give it a slightly grippy but slippery surface.
it walks pretty well on carpet too. uneven surfaces are a nightmare though! better now that i have removed those big feet.
yes this plays many factors, but i do believe one could get the servos to slow there rotation once foot makes contact with the ground allowing for smoother landing. in theory.
You’ve been pretty busy lateley! Especially for a guy who isn’t really “handy” at the moment. LOL
I didn’t read trough the whole post but I like the “cat” style quad setup. Are you using the phoenix code as base? Are you shifting the body (COG)?
hi xan, i wondered how long it would be before you found it.
yes its running on v1.3. iv posted the code on page 7 download/file.php?id=642
yes after the stalker quadruped robot i was determined to get a quad walking. and finally i did. im not using any body shift. its purely done by a well calculated gait.
and of course 'a well executed code thanks to you.
i am waiting to get a better camera before i do a final video.
i have made a few improvements and zenta has helped with some of this.