Industrial Robot Arm Becomes Giant Catapult
wintersynth brings us a story about a group of enthusiasts who made a catapult out of a 2,800lb industrial robot arm. They used it to launch bowling balls, fireballs, and cans of beer toward a stationary target, and they controlled the catapult's aim with a graphical UI on a laptop.
"I wanted to be able to control the rotation of the robot so we could aim the robot from the laptop, but I quickly realized that since the desert is so flat, we could do some basic ranging on the target too. I also wanted the targeting to be overlaid in 3d over a photograph of the target area. The software needed to control the robot like an MMO or RTS game. I suspect that video games, in general, have some of the most optimal control interfaces. I wanted to try a control scheme similar to the area effect spell targeting in World of Warcraft."
There were non-counterweight trebuchets as well, called "traction" trebuchets. Instead of a counterweight you had a number of people tugging on ropes. I had one based on this model built for me for SCA combat as the result of a siege engine competition (Stormhold) some years ago. 60-90 metre throws with a cargo of softballs was customary with a 6 metre composite rattan arm. One advantage of a traction trebuchet is it's more mobile as you don't need to score or drag a tonne or so of counterweight along to the launch site.
So to stay on topic, I think you could call the robot arm a form of trebuchet. I've not seen onagers with slings in my researches though, will look for that. Onagers btw were so named because of the bucking motion they make, mitigated by curved ends of their foundation rails. Onager = Donkey in Latin. They were also called "rocking donkeys".
And another name for Ballista could be "ZOMG Look at the size of that effing crossbow!". They didn't always use rocks, some of them used mucking great iron bolts.
Do not mock my vision of impractical footwear
Okay, I have to admit, we felt pretty bad about this "renting" tactic until we actually tried the camera. It was hands-down the worst HD camera I have ever used. I mean seriously, it had all sorts of proprietary software with weird codecs so that the footage was extremely difficult to transcode at high resolution.
I felt absolutely no remorse returning that thing. I know, that still doesn't make it right, because we didn't know that going into it. But I hope it is at least a mitigating factor. Plus, I give Fry's tons of (non-"rented") business, and their awesome return policy is a big part of the reason.
But that incident, among others, spurred work to develop collision detection. They finally got some software running on the DSPs that'd estimate what the current to the motors should be, and measure what it actually was; too big a difference and the robot would halt. And then comes the fun part...
I got to test it.
For six months, my paid job was to take huge industrial robots and bang them into things.
I'm pure software now, and it's fun and pays better... but I still think about those days with fondness.
PHEM - party like it's 1997-2003!
The way you protect workers from getting killed by a robot (and these things are way stronger than you think, even after seeing it fling rocks) is to put up light curtains around the robot.
The OSHA safe stopping distance calculation is used to prove that the hazardous motion will stop in the time it takes the person to traverse the light curtain and come into contact with the equipment.
So, the safety folks find the robot with the biggest, fastest moving load on the line--the floorpan skin transfer robot. A floorpan skin is basically a sixty-pound razor blade.
The end effector held onto the floorpan skin with suction cups, which are a cost-effective and reliable method for the process.
The robot guys set up a test, where they got all 6 axes of the robot moving in such a manner that the end effector achieved its maximum possible speed.
Not something you'd normally do, but a worst-case scenario for use as safety systems challenge.
We all wanted to see this robot haul ass, so the safety folks had us all standing back...
Robot dude picked up the TP and initiated the path at 100% speed...
Somebody waited for the arm to get to full extension and speed...and stuck their hand into the light curtain.
The robot stopped almost instantly--well within the expected stopping distance.
No way that person would have been injured by the robot.
The skin (remember the sixty-pound razor blade) stopped a couple bays over.
Hard clamps were added to the end effector and the test was repeated with improved results.
"Reality is that which, when you stop believing in it, it doesn't go away." - Philip K. Dick