This Week in TechnoloGeez — Robotics
By Mark Devlin
October 27, 2009
BigDog’s Biped Offspring
Probably the most interesting of possible tech topics this week is robotics.
(Really stunning robotics in manufacturing? I’m still waiting. Listening. Watching. Holding my breath, even.)
It’s tough to say exactly which bit of robotics news is the most interesting, however.
Remember the BigDog? …
With Japanese humanoid robots often having President Gerald Ford-ish propensities to just, well, fall over…
…the BigDog has stood (fast) as a great robotic achievement not only because of hardware, but the all-important software behind it all. Look at the above video again. BigDog doesn’t fall over, but snarls and growls “Go 'head. Try to knock me down.” Plus, with impressive load carrying capability, it’s got real-world applications.
BigDog’s developer, Boston Dynamics, has moved from quadruped to biped with the Petman…
According to this piece over at Wired’s Danger Room…
“Petman will balance itself and move freely; walking, crawling and doing a variety of suit-stressing calisthenics during exposure to chemical warfare agents,” the company promises. “Petman will also simulate human physiology within the protective suit by controlling temperature, humidity and sweating when necessary, all to provide realistic test conditions.”
Yo. Boston Dynamics. Can we get one of these things moving inventory, maybe? Or schlepping equipment on a construction site?
Paradise by the Dashboard Servo
From four legs to two, and now to four wheels, here we have Shelley the autonomous Audi, from Stanford University…
…according to this piece over on BotJunkie.com.
Autonomous cars aren’t all that new, and with DARPA Challenges, it doesn’t seem a stretch to program and rig a car—albeit a really interesting and mildly sexy one—to run in circles on a wide-open salt flat (or, what appears to be so). The author of the BotJunkie piece says, though, “Remember that the point of this car is to test the limits of autonomous driving.” We’ll see.
Watch the video, especially if you hold an interest in rally racing or the Pikes Peak Hill Climb. That’s where Shelley’s headed next year. Impressed? If Shelley turns in a good time, there’ll be reason to be. That’ll be some interesting sensing, programming, and control over a wide world of vehicle dynamics.
If you prefer things to move a bit more slowly, check out Stanford’s other work with a VW that parks itself…
Impressive? Not to me, but I always park in the middle of nowhere to avoid runaway shopping carts, so maybe my perspective’s off. Make it power-slide into the parking space a la Starsky and Hutch. Rig the doors with sensors and adjustable limiters so the morons going along for the ride can’t ding the sheetmetal of flanked cars. Those would be impressive. But, looking at the positive, there’s at least there’s some compelling sensing and geometry in the slow-mo parking maneuver.
Autonomous, robotic cars? With most humans eating, applying makeup, watching TV, playing video games, screaming at the kids or talk radio, and making cell calls while seeming to drive, it seems we are in desperate need of robotic assistance.
And, um, Stanford? Have you noticed, for example, what Seegrid Corp. has already accomplished in the industrial world? You might want to check it out. I don’t think the ultra-slick Seegrid technology would have a problem parking.
Swarmbot Revolt Closer
When we think of robots, foes, and robotic foes, we might think of Gort, The Terminator, Sonny, or C3P0. (Relatively large form factors.) Or, Optimus Prime. (Really large form factor.) High to insane power requirements. Heavy metal and motors and hydraulics to move around. Intense programming requirements. High maintenance.
Maybe, just maybe, thinking that big isn’t the way to go. After all, armies could be as easily defeated by something microscopic as they could be by tanks or Sidewinders. Certainly, entire towns and regions have been laid waste by locusts—insects without a lot of processing power but one kick-butt swarming mechanism.
What about something small, but not microscopic? (Heck, even the “kinda cute” Tooth Fairies of the [allegedly] fictional Hellboy II movie illustrated that swarms—living or robotic—can wreak nasty, particularly gruesome levels of havoc [Click on this link at your own risk...it’s a safe link at MSN, however.]).
What about...hmmm...let’s say sorta billiard ball-sized bots that—in reality—have been programmed to study swarm intelligence. Take a look...
According to this recent article over at h+ magazine...
Each bot had an initial built-in attraction to a “food” object, aversion to a “poison” object, and a randomly-generated set of parameters –- their “genomes” –- to define the way they move, process sensory information, and flash their blue lights.
The genomes of the bots that found food and avoided poison were recombined by researchers, mimicking biological natural selection. To create a next generation bot, traits were combined and randomized to mimic biological mating and mutation. First simulated in software before using actual bots, five hundred generations were evolved this way with different selective pressures by roboticists and biologists at the Ecole Polytechnique Fédérale de Lausanne in Switzerland in 2007.
Check out the article for some mind-bending reality tech ending on a provocative note...
Are deceptive bots the real threat? Probably no more so than humans who lie and cheat. Perhaps bots will simply end up evolving on their own into something else again...
Huh. Um, enjoy your week.