I actually just bought a 2.0ghz laptop to run the wireless Visual processing for the face tracking head and to transmit the sequencer commands. I needed it for an animatronics project I’m doing for an advertising company anyways.
So we have a brain for the scorpion now.
Your tail is a go as far as I’m concerned. It’s a great idea. The servos didn’t show up today so hopefully tomorrow. Once the legs are on, we can start on the tail and stuff.
After losing most of my hair trying to program the Atom Pro, I have a PS2 controlled walking program working. Sort of. One middle legs goes out straight when you push the analog sticks too far. And the tibias go out to far sometimes.
For some reason the SSC-32 offsets aren’t perfect on one leg either.
I’m cleaning up the wiring and building some lipos to sit on it so it will be ready soon.
There she is. Not a great video but a video none the less.
There are a few legs out of place and it freaks out every now and then. Don’t know why. And the Tibias need to be offset inward more. The legs spread out randomly for some reason. And at the end when the battery dies, it looks like it’s squirming.
It seems the HS-77 low pros are barely cutting it. However, after 10 minutes of walking, none of the servos were hot.
I’m using the program kurte was nice enough to convert to inline hex and atom pro. I used it on the Giant Hexapod.
I would love to figure out how to use zenta’s ripple gait and load it onto the atom. But I’m still trying to figure out how to properly convert the .csv file and to actually use the spreadsheet.
I also did not set any of the angles. Just the SSC-32 offsets to make everything line up. And I modified the program for the leg dimensions I’m using.
Maybe zenta could bless us with a .csv file for our leg dimensions and the ripple gait.
Our hexapods don’t have this randomness. So… I noticed you are using a tethered controller. I’m assuming it may be a sony brand. They have the worst joystick performance compared to many of the 3rd party ones. I dunno why. But anyway Kurts code is based on Laurents code, which was fine tuned for our wireless controllers. I’m referring to the actual available range from the joysticks. They are not all exactly the same, and some parts of the code can be effected if a value is a little out of range. Laurent is hoping to have his Phoenix walking under PowerPod control this weekend! If I remember correctly he’s usoing the Pro for this. So we should have something to offer soon. It looks pretty cool, nice work.
Yup, it’s a sony and an old one too. I noticed the joysticks can sometimes provide undesired ranges. That explains it then. I have a madcat I’ll try.
I was about to ask about Atom Pro support on the Powerpod. That would be really cool if he had a pro code for a hex walk. Especially a ripple gait like the phoenix uses.
i was wondering what it was doing at the end, , i thought it maybe freaked out, or maybe it was like, a “squirm” program u put in for some reason…then i read the rest of ur post
again, thing looks awesome, cant wait to see it done 8)
Nice work! And cool video!
If you want me to make a simple ripple gait, one seq forward and one seq backward and some body movements for a test, I need to know the leg dimensions: HipH-HipV, femur length, tibia length, lowest part of body. Exact body dimension, meaning X-Y coordinates of coxa axis relative to centre of body. Center of body => X=0 and Y=0. All dimensions in mm (please ).
If you have a project in SEQ for your scorpion it would be useful to have a export (both the .shp and .csv file).
Anyway its important to setup your bot in the SSC32-configuration correctly.
Send me a PM and I’ll give you my email adr, and I can have yours.
I’m hoping I can find a way to load your sequences into the atom to respond to certain PS2 inputs. I’m not sure how I would do this though. All I would like is to have a ripple gait on the scorpion without having a computer hooked up.
I changed the ps2 to a wired Madcatz and it works way better. I still would prefer a ripple gait over the tripod though.
I don’t know near enough to use the GP sequencer so we’ll have to wait until Lareunt releases the new powerpod that will have ripple gaits and support for the atom pro. zenta sent me an awesome ripple gait from his PEP program and I can convert that to a basic program. But I wanted the scorpion to be separate from a computer.
I may either have to go wireless for the serial port and control him from SEQ or learn how to get the program working in the atom pro with PS2 controls.
Either way, here is the current progress on the tail and head. The head is on a pan/tilt and has a color camera and Sharp IR Sensor.
The tail consists of 14 Ball Bearing joints with 7 DOF. The tip isn’t done yet and I haven’t rigged a servo to control it yet. But the springs are there and it’s ready to go. It looks really cool when you rock the body back and forth. The tail extends outward and contracts back in on it’s own. I think it might look pretty cool just swaying freely as it walks.
Here are the pics.
The tail only adds 6 oz’s to the hex for a total weight of 5 lbs 1.1oz’s.
That’s including all parts, 6 HS-322’s, 6 HS-645’s, 6 HS-77 Low Pros, 2 head servos, the tail, a 3 cell lipo logic battery and a 5 cell 4400mAh sub-C servo battery.
Looks cool. I have my Round hex somewhat torn up right now to update the SSC-32 to V2 plus memory, plus a DPDT switch to allow me to have it talk to SEQ.
If I get a chance I might try hacking a version of the code out that will allow it to run a sequence. Maybe if you sent me the sequence I could try putting it into my hex and see if it would at least call the right sequences when some of the switches are pushed.
Note: not sure how much time I will have to play with it as the new Pro software is fixed and I can now continue hacking on my rover!!!
Drop me your email in a PM and I’ll send the 2 files for the PEP spreadsheet, the SEQ project and the basic program. You can tinker with them if you like.