Bing 3

[DerekZahn]

One more picture of the leg supports to show how the wiring gets routed:



Bigger: http://happyrobots.com/bing3f2.jpg

[Joe]

Derek, how did you make those cool plastic parts? Those aren't from the milling machine, are they?

Thanks,
— Joe

[DerekZahn]

Yes, the beveled edge is cut with something called a "chamfer bit". They were kind of challenging to make, but they came out well.

[Robo1]

Loving the design, just wondering how big this robot will be. I've been having problems designing my bot and keeping the size thus weight down. for my ankle/hip it's 70 x 90 x 60 which is big just wondering how big you'res is.

bren

[rockweasel]

Just.....beautiful! Looking forward to the next stage.

[DerekZahn]

Loving the design, just wondering how big this robot will be. I've been having problems designing my bot and keeping the size thus weight down. for my ankle/hip it's 70 x 90 x 60 which is big just wondering how big you'res is.

bren

I'm not completely sure what the total height will be as I have not finalized the design of the torso yet, and whatever head gets stuck on will impact the height. The current sketch has a height of 32 cm not counting the head and standing at full height. The little pods are oddly-shaped making their size a bit odd to measure, but the ankle pod (and also the upper leg pod which is identical) without the wiring cover and bearing attachments is 65x45x45. With all the doodads bolted on it's 70x50x50. The knee pod is 50x50x45.

I don't know what the weight will be but my guess is "heavier than I was hoping". I might have to go through and lighten things up a bit. Keeping the pieces as small as possible was one of my goals but looking back I would have added a little size to make the wiring a little easier, the clearances are pretty tight which makes it time consuming to get it all in place and replacing servos when they fail will not be pleasant.

How big are the DX-117 servos?

[DerekZahn]

I still have some hope of finishing the legs today, I have a bunch of stuff to do on the upper leg pods still, but it's all stuff I've done before for the ankles. Here's a nice picture of the current state though:



bigger: http://happyrobots.com/bing3g2.jpg

[Robo1]

The DX-117 are bigger then I wanted as there 40 x 30 x 45 not including the servo horn that sticks out 8 mm

As you can see it's roughly the same size but wider.

my robot will probably stand at about 50 cm but I'm trying to make the design smaller so less mass.

bren

[rep001]

WOW !
i have only just tuned in to this thread.What a fantastic achievement.I am still working with code that is available to me and you are building from nothing.I bow and tip my hat to you sir !

8O

[DerekZahn]

Ok, after a struggle, the legs are ready:



bigger: http://happyrobots.com/bing3h2.jpg

Next I'll keep working on the torso, get the design of the arms nailed down, and figure out the attachment for the legs to the torso.

I also decided that having one battery cell in each foot was kind of a dumb idea because it makes the wiring too complicated. So I'll be redoing the feet sometime in the next few weeks.

[Robo1]

What about having two cells in each foot then you can have twis the power and run them in parallel.

bren

[DerekZahn]

Thanks, that's a pretty good idea; it would certainly help with some of the wiring issues.

However, making the feet shorter and lighter appeals to me so I think I'll move the pack to the torso instead... because I'm planning to experiment with dynamic balance and walking I don't mind moving the center of gravity upwards a little bit (trying to keep the CoG low was the primary reason for putting the battery in the feet in the first place, but I have had a change of heart on this topic).

As an added benefit, it will be easier to swap batteries, which could come in handy, and I get to move the ankle joints closer to the ground, which was a design goal for me anyway.

[Joe]

because I'm planning to experiment with dynamic balance and walking I don't mind moving the center of gravity upwards a little bit (trying to keep the CoG low was the primary reason for putting the battery in the feet in the first place, but I have had a change of heart on this topic).

Can you elaborate on this? I'm still thinking that keeping the CoG low (e.g. by putting batteries in the feet) makes balance much easier, as well as simplifying things like getting up from a seated or lying position. I've also noticed that most consumer robotics (e.g. Robosapien and various animated Elmos) locate the batteries in the feet, presumably for the same reason.

But I'm eager to learn new ways of looking at things... so can you elaborate on why you no longer feel a low CoG is worthwhile?

Thanks,
— Joe

[DerekZahn]

For the same reason you don't have 40lb feet yourself -- the more maneuverable the legs are the better.

But beyond that it's not so much that a higher CoG is always better, just that lowering it is not really advantageous if you have sufficient ability to balance dynamically. Most robots don't, but it's one of the things I would like to try to accomplish.

As a cool example, consider the famous MIT biped:

http://www.ai.mit.edu/projects/leglab/r ... biped.html

[JonHylands]

Hey Derek,

Long time no see

Bing looks way cool. I'm doing a biped as well now, which you can see at http://www.bioloid.info/tiki/tiki-index.php?page=MicroRaptor.

I'm curious about your foot sensors - how do those work for touch on general surfaces? I thought those were only good for skin-type touching...

- Jon