Let's Make Robots!

812-R7 Biped

Walks in a circle (no, not on purpose)

*Update 2010 Jan. 26* 
I figured out one of the problems with R7 was there was too much flexibility in its feet. I added several polymorph goobers around the edge and against the brackets to strengthen them up a bit. Now it doesn't fall flat on its face anymore! I even managed to get some (crappy cellphone) video so you can have a look at R7 taking its first (mostly) successful steps. Definitely got quite a bit of work to do on the programming...

 

I've been wanting a biped for quite some time now. With the holiday acquisition of half-a-dozen micro servos and a kilogram of polymorph I've finally got this project underway. I won't write a whole helluva lot here right now, y'all will have to mostly settle for a few captioned photos.

Most of the polymorph is rolled flat using ~1/16 inch thick tongue depressors as guides. Then cut, sliced, bent, melted, rebent and stuck in a U shape. The pivots are 22g copper wire, not the most robust choice for hinge material.

Currently there are ten micro servos. I only had 8 of the Dagu servos, so the two black ones (shoulder rotate) are pirated from my helicopter.

The brain is an ATMega168 @ 16MHz running the Arduino bootloader for easy programming. The board is a Gadget Gangster half-size project board. 12 of the 'mega's digital pins are brought out to three-pin "servo style" headers, with a strip of female headers above. Obviously only ten are currently in use by servos. D13 has a bright blue LED for debugging just plugged into the female headers. None of the analog pins are brought out yet. Top left is a 1x6 right-angle header for the programming cable. There's a 5V LDO regulator to provide power to the chip from 4 AA or AAA batteries (alkaline, totalling ~6V). The servos get their power straight from the batteries.

In all the whole thing is quite wobbly. Combined with the fact that I don't have a clue what I'm doing, R7 manages to get about 1.5 steps before tipping over on its face. I'm fairly certain a redesign of the feet and ankles is my next step before working very hard on a walking sequence. I have managed to get it to move to standing on one foot though. That felt like at least a small win ;)

For now I need to get back to working on LMR again, so it may be a little bit before I do more on this. Hopefully I'll come up with a proper name for it too.

 

P1020691.jpg
Hips and Knees - front

P1020694.jpg
Right knee and ankle - front

P1020695.jpg
closer view of right ankle - front

P1020697.jpg
knees - front

P1020698.jpg
hips - front

P1020700.jpg
left shoulder and elbow - front

P1020707.jpg
left elbow

P1020712.jpg
back

P1020713.jpg
legs - back

P1020715.jpg
left ankle

P1020719.jpg
left leg. Also shows a little bit of the Erector stand I knocked together to hang the bot for programming

P1020722.jpg
Brain "backpack"

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Wohooo, it's walking!! how nice, Del!! Respect!

With the Recent snowy posts we have been having, looks like you have a good contender for the "Abominable Snowman".

(-: PolyMorph snowshoes included :-)

Watching the video has increased my respect for bipeds - the "Cool" guy is great mover.

you got an amazing work there! Congrats Rudolph!!!  ;)
I wonder how you will program a simple walking sequence with so many servos to control!!

keep it up  ;)

Hi Ru, missed you when you made this :D (This robot was made because Rudolph did not have any internet-connection, so he did not have me on he's back to go on about LMR development :D)

Dolph, are you not going to give it any sense of orientation / tilt ?

Eventually it'll have some kind of "I fell over" sensor. It will likely just end up being a ball-in-cage thing to keep it cheap and simple. Someday I may try an accelerometer too if I ever pick one up.

since you've put such a huge amount of work into the design of the servo brackets, any chance that you might post a plan of them? perhaps you could encourage dagu to mass-produce some - since it looks like you've used some dagu-type mini servos there...?

pleasepleasepleasepleasepleaseplease....etc. 

Once I get some kinks worked out, and figure out a good (i.e. repeatable) way of building them I'll provide more details about how to make them.

Eight of the ten servos are from Dagu. The two non-dagu ones are the shoulder rotate ones which aren't using the brackets anyway. I would be thrilled if Dagu likes them enough to produce them. And I'd be nothing less than ecstatic if Dagu liked it well enough to make a micro-servo biped kit  ;)

I have actually had our engineers working on the brackets but other projects keep taking priority plus it takes money to have the dies made. As for the servo brackets, did you read my walkthrough on making such brackets. I even made a tripod to take video.

I remember seeing some cad drawings you'd put up somewhere about a mini servo bracket that you were trying to get produced. These ones are rather different though and (I believe) allow for a narrower overall "footprint" (can't think of the right word that I'm after).

Yes, your walkthrough helped me quite a bit in learning how to work with this stuff. So far I've been unable to locate tempered glass placemats/cuttingboards that aren't textured in some silly way. I quite like the idea of smooshing the poly flat rather than rolling it, I just can't stand the thought of all my poly sheets having a pebble finish ;)

Emuller have a point though; There must be only a few basic brackets necessary, so one can put servos together in 0 or 90 degrees, on 2 axels. Then any sort of strange stuff could be made, by putting servos together.

The need is there all the time!