Robot Army : Kinect Success!

deltaDo

Today was cold. I drove to Mark’s with frost caked on my car from the morning dew. BRR. Once I arrived, I drank a million cups of coffee and attempted to get some work done on my spread sheet… but let’s be honest, everything I did today was boring (prices and PR), and everything Mark did was amazing. So lets focus on his achievements.

Ever since Saturday Mark and I have both been trying to figure out how to get the Kinect communicating with some servo motors. It seems that all the example code on the internet no longer compiles because of some issue between the new processing, java, and Arduino. In short, Mark began from scratch and at last today, typed in the magic numbers that made everything work.

The Kinect isn’t controlling actual delta robot kinematics however. For today, the triumph is in having mapped a motor to each of the x, y and z-axis of your hand. That’s right, if you wave your hand around… the delta robot will move according to where it is in 3D space. It isn’t very intuitive looking yet, but by golly it’s a damn good start. As soon as Mark figured this out we cracked open some beers and filmed this video, calling it a day. We’ll start on the more complex stuff Friday. Good job Mark. You’re a wizard. =]

 

 

Robot Army : Ditching the EEG

Print

After a jolly romp to Quick Care to get antibiotics for my fluffy sinus, I ‘m in bed again writing emails. This PR stuff so far is feeling a lot like standing on my tip toes while shouting through a cardboard tube. >.<

I did head down to SYN Shop yesterday for my first late night hang out since this spring. One of Mark’s old friends from Sun, Tsutomu, was in town for LDI (the lighting trade show) and I got to pick his brain while I was there. He was a great wealth of knowledge and gave some good feedback about me and Mark’s collaborative work as well as my light installation. Long story short, he persuaded me from using an EEG to control the robots like I was planning to. He voiced everything I had already assumed about the reality of using neural input to control anything electronic. Brain noise is noisy.

I thought for the past couple years that I could use something on the market like the Emotiv Epoch to easily decode the mess of signals coming from my brain and implement them as data inputs. In spite of my excitement about using brainwaves in my art, I held off from buying their device for a couple of reasons. One of them being that in the whole world of hacking, I hadn’t found one other example of someone successfully using an emotiv or anything else like it to do something substantial. The few hacks I did run across seemed more or less like slop turning an LED on and off by chance. There was nothing to convince me that the data people were using was reliable or consistent.

Turns out I was right. EEGs aren’t reliable or consistent. Any sensor used outside the brain is susceptible to picking up huge amounts of chatter from who knows where. Devices like the Epoch give you a tool- but it is more or less up to you as a developer to figure out how to read the input and then decide what (if) you can do anything useful with it. ::sigh::

The solution? Who knows. I have to refocus on the more important matters at hand… like raising funds to build the army! Back to that.

So in other news, I sent Mark’s second revision of the Delta brain off to OSH Park last night – and I was pleased to receive an email this morning already informing me that the panel I was assigned to went to fabrication (cheers). Yesterday,  I also ordered some cards for Pawel and Suz to hand out at their robotics conference in Florida next month. They’ll be taking Santo along with them as a demo…. which means very soon Mark will load him up with code to make him charming and winsome as ever.

Tomorrow morning, more progress and another video.

Robot Army : Cuddling my Spreadsheet

photo_1

Today, I find myself wanting desperately to feel as though I’m doing something right. I think it’s likely all the rain we’ve had that’s tainting my mood, but I’m discouraged. I’ve been adding things to my bill of materials and the proportions are seeming less than hopeful in some areas.

Now I need to start shaving cost off things that are needlessly spendy. I guess this is where I learn to be inventive and shrewd all at once.

It might just be my lack of experience in doing PR related stuff- but I’m stuck again with the getting the word out part. Maybe I’m not begging loudly enough. I’ve never been all that good at asking for attention. So this is another hurdle to overcome.

This morning I did mock ups of several different variations Mark came up with for my delta’s arms. We foresee the cups that hold onto the ball bearings wearing out over time, so we’re trying to come up with a solution. One idea was to have a slit down the center of the U-joint that some hardware would compress together so that the tension on the ball could be adjustable :

photo_3

The other thing Mark suggested was that I ditch the cup entirely. He said  it would be too difficult to mill if I decided to have my pieces machined. His idea to replace it was a simple square peg that had a “+” sign cut into it… leaving four small notches that would then grip the ball joint. I gave that a try too.

photo-2

The compression design worked well. It’s a good idea, but if I go with this type of arm, I’ll have to add another 12 pieces of hardware, plus 12 washers and nuts per kit…

The replacement for the cup worked in theory- but the ball was apt to pop out left, right or any of the directions it didn’t have material forcing it in place. So maybe not such a good idea after all.

Other than that… I’m attending a hardware startup meetup tonight with Jeff and Mark. We should all learn something from it. Tomorrow is Friday and I’m back over at Mark’s to finish and send off Rev 2 of the delta brain to be fabbed at OSH Park. Small steps. Chin high.