Archive for April, 2010

We’re trying something else!

Posted in robotics, sound on April 27, 2010 by raphaelabrams

Now we’re getting somewhere! We dropped the whole mechanical approach to gathering facial movement data. Instead we’re trying to use the sounds themselves as the control.

Step one is to attach a microphone. I found a nice little package with a built in preamp and low impedance output. Oh, and it’s really really small:

See the rectangle with the black dot? That’s it. The rest of that is mostly hot glue and wires that I applied, just so we could get a grip on it. It’s about 1/10″ thick. Thankfully it only has four solder pads so it’s reasonably easy to deal with despite its size.

It’s a Sisonic from Knowles. I think it’s neat-o!

Great! Now we have a decent analog voltage to read into one of the analog-to-digital channels of the AVR chip. All we need to do is read that signal, look for peaks and troughs in the waveform and measure their amplitude.

UPDATE: NOISE! ARGH! STILL TOO MUCH NOISE!!!!

Servo control board!

Posted in robotics, servos, Uncategorized on April 27, 2010 by raphaelabrams

Here’s a circuit board I designed for robots and animatronics. It’s an AVR chip with the Arduino firmware flashed onto it. It has eight servo hookups on it, and a handful of analog and digital inputs. It’s tiny and reliable. I’ll be using it as the motherboard for this project. You may know it as the Twitchie.

In this image is the battery compartment, two servos, a potentiometer with a wad of red duct tape on it and the circuit board. Just in case you’re wondering, it’s running at 16MHz, has voltage regulation for the chip, direct power connections between the servos and the battery case (servos like a few more volts than microcontrollers to really get them going) and is clad in a very dashing shiny black solder mask.

First attempt at lip synching is… interesting.

Posted in sculpting, silicone on April 27, 2010 by raphaelabrams

We have to make the mouth of our giant puppet move in sync with the mouth of the dancer inside. We tried making a silicone mask to hold an array of sensors to read the movement of her face. It doesn’t really get us what we’re after. We want the mouth to open and close on each syllable, but real people don’t do that when they talk. In reality there are a vast number of movements to read, most of which are in the mouth or throat. Reading these mechanically is not just difficult, it’s really uncomfortable!

Anyway, below are some photos of the rig we tried to use to read the chin movements.

First we made a plaster cast of one of the locals. We then made a clay covering, which is used to make a further plaster cast.  Finally these two casts are used as a mold for making silicone chin harnesses.

The yellow things are silicone casts. There are two human chin harnesses and one giant sized puppet mouth, all made from very soft silicone.

That big mouth has a self satisfied, smug quality doesn’t it? Well, don’t worry! We beat that out of it with one super powerful servo and a good bit of crazy glue.

Here’s a chin harness. We’re trying to glue in a flex sensor. It works, sort of. OK, hardly. (You may notice that we have the sensor bent the wrong way. We flipped it and still got vague results.)

The other bit we tried was a simple potentiometer mounted at the hinge of the jaw, with a rod going down to the chin. Again, it works, sort of.

We’re going to try something else!

Wiggly minions, I bring you LIFE!

Posted in analog, arduino, robotics, servos, sound, Uncategorized, update on April 26, 2010 by raphaelabrams

Hi folks! This is Raphael Abrams here! I’m an artist living in Brooklyn, doing all kinds of odd skilled labor to make a living. I teach electronics at NYC Resistor (I’m a co-founder), design open source electronic kits, do contract electrical engineering and when I get the chance I try to make art.

This blog will serve as a record of my new adventures in animatronics and puppetry. I’ll be sharing what I learn, and I hope you might be able to find something useful!