I really like the way it moves. Specially at the last part of the video, where it is "running". One day, with 6 small geared motors.. I have to make something like that! :D
Liked the idea of having the driveshafts spinning faster when there weren’t any load on them, then slowing down to the tempo set by the remote control as soon as the “foot” met friction/ground.
I think that with this many legs, (and the spin-faster-when-in-the-air method,) they probably dont need any more logic. Just normal tank style steering. Oh, I’m sure there are more programming involved, but I think that it is possible. Theoretically, the rotating feet are just replacements for rotating wheel, very theoretically…
Building one of these “with 6 small geared motors” you say… Are we talking GM10 kind of small motors or Tamaya 7,2volts kind of small motors?
at about 1:00 in the mud pit, you can see one side being driven backwards while the side closest to the camera is not. This makes it likely to belive that it is steered like a tank…
at 1:20 it looks like the legs are springs, because the steps down toward the water sure looks spunky…
During the running part at the end the feet are actually spinning with the same speed during the entire 360 degrees, (downloaded and ran it on low speed in VLC) so that’s just an easy matter of making a six wheeled vehicle and switching the wheels for banana shaped, rubber tipped, spring loads
However - look at the part with the stairs. Is that automated? It changes motion… And the part where it has fallen off the rocks, feeling the ground?
It may be low-tech - but it could also be a matter of constantly checking ground, thinking of "what do we want", and then spinning accordingly, based on pure math.
Any other company I might have thought it was just low tech, hard-coded stuff… But seing their "Big Dog"… well… these guys are capable of doing some good leg-calc on the fly
On the other hand; It IS remote controlled… so shifting to "stairs-mode" would be the easiest way.
I wonder if there is a really low tech way of detecting the spins if I was going to make a very small one just for fun…
Yes, the RHex (what’s with the dog-fetish anyway?) is without a doubt more sophisticated than i suggest. What I’m saying is that it is possible to fake some of the functions, with easier work-arounds
As for “a really low tech way of detecting the spins”: the guts of an old ball mouse (not laser) seems like a good place to start. There are several different hacks on internet utilizing these encoder. It might not be exstremly low tech, but they’re small…
Lots of PICs have comparators. Do any of the picaxe chips you use have comparator inputs? If so, you could do a voltage drop detection. You could even use an analog input. You’d need one per motor…
But I wouln’t bother. I don’t think it’s critical that the feet go faster in the air. Just make them go flat out!!
There seems to be about a dozen institutions and companies involved in this project. I’ve seen sandbox innovation’s robots up close at a conference a few months ago - and it’s pretty much identical (I also know they have relations with UPenn, one of the partners mentioned in the video just posted. Here’s another: