Web Concepts

Posts Tagged ‘robotics’

The main problem with MAVs has to do with the way they respond (or don’t respond) to dynamic environments--things like shifting or gusting winds, moving bodies, and other variables that have to be accounted for in realtime. MAVs are tiny, so there’s not a lot of space for computing assets or sensor payloads, and that leads to a sort of intractable problem: how can engineers make these things smaller and more capable while also adding increased situational awareness and better in-flight processing?

When facing a tough problem like this a little biomimicry never hurts, and that’s exactly where the Pentagon is looking with its recent contracts. If two research stipends recently handed down are any indication, the micro-drones of the future may have tiny hair-like sensors all over their bodies and big, compound eyes.

The cilia-like hairs will serve to keep the drones’ hovering and flight stable by sensing changes in air flow at the tiniest levels. That means the drone could sense a wind gust coming shortly before it arrives, allowing it to compensate for the change in circumstance. It would also aid in maintaining overall stability during flight, as the MAVs central processor would possess a constant awareness of--and the ability to manipulate--the boundary flow layer of air surrounding the drone as it hovers and flies.

The bug-like compound eyes would similarly help MAVs navigate in cluttered spaces by increasing the amount of visual data available to the drones’ processors. An on-board minicomputer would process images in realtime, using those visual cues to automatically avoid obstacles and navigate cleanly and efficiently.

[Danger Room]

The footage comes via an event at Southmead Hospital in the U.K. aimed at raising awareness of men’s cancers. To show just how effective Da Vinci can be in the operating room, urology fellow Ramesh Thurairaja grabbed the sticks and delicately showed a grape just what Da Vinci is capable of.

There are more than 1,000 Da Vinci robots worldwide, and this particular robot has performed 450 prostate cancer removals alone. No man wants to think of his grapes anywhere near the forceful hands of a massive multi-armed machine, but this demo shows just how magnificently precise and steady-handed our robot surgeons can be.

[WIRED UK]

Fingerprints provide a unique identifier and a better means to hold on to objects, but they also shape the ways we sense and perceive the world around us. When we touch something, the ridges alter the vibrations moving through our skin such that nerve endings can better receive them. This serves as a kind of signal processing that allows the skin in our fingertips to provide richer information to our central nervous system than skin on other parts of the body.

For robots, who generally do a lot of processing in a single central processing unit, this distributed signal processing represents a distinct advantage. If robotic sensing surfaces could do more signal processing in the surfaces themselves, it would save on the amount of processing taking place in the CPU, leaving more room there for other kinds of computing.

So the National University team created a robotic touch sensor out of four force-sensitive sensors on a single four-millimeter square plate. Using a thin, flat plastic sheet covering the sensors, they recorded a series of force/touch measurements. They then repeated the measurements, this time covering the touch sensors with a ridged plastic sheet. They found that the ridged measurements where far richer in touch data, even helping the sensor identify the shape of the object being touched.

That all happens right at the “fingertips,” saving the central processor the trouble. And it represents a nice distributed solution that could greatly enhance robotic object recognition. It also represents a nice piece of biomimicry. Read the paper at arXiv.

[Technology Review]

We know that various research and academic institutions are working on robot autonomy (regular readers see stories and videos of these autonomous ‘bots right here on PopSci all the time), but what’s a bit mind-blowing is just how far along some of this technology is. At Fort Benning, a team of Georgia Tech computer scientists is helping the military demonstrate software that can autonomously--without a shred of human input--acquire and make life or death decisions about targets on the ground.

That is, the only thing that’s missing is the capability to fire. Add that, and you’ve got a killer robot.

Of course, these are just demonstrations (for now). But they create a blueprint for the inevitable future of warfare: when time is critical and running decisions up the chain isn’t feasible, software will make key decisions about what constitutes a target, what falls within the bounds of the “rules of war,” and whether or not it’s safe to commence firing. If a program can satisfy whatever requirements have been seeded in its coding, then it’s bombs away.

It all sounds a bit Skynet, but it’s moving forward at a rapid pace within the U.S. military, driven both by need (putting fewer human lives in harm’s way is obviously preferable) and that Cold War-esque mentality that if America isn’t at the front of autonomous warfare then it can only be behind. That sentiment is not entirely misplaced: South Korea has already deployed semi-autonomous armed robotic systems along the demilitarized zone bordering North Korea, and the Chinese have a dog in the hunt for autonomous weapons systems as well.

So what is the state of “lethal autonomy?” To put a number on it, it’s at least a decade (probably more) away from becoming battlefield reality. I sat in on a lecture at last month’s AUVSI unmanned robotics conference titled “Armed and Autonomous” where the focus was on the idea of deploying armed UAVs into contested airspace--using unmanned planes to deliver surface-to-air and air-to-air weapons in areas where anti-air defenses are still intact.

What might surprise many is that the computer programs necessary to evade air defenses and execute these kinds of missions autonomously already exist. The backbone technology is there, we just don’t trust it enough to actually deploy it. The idea of unleashing armed and autonomous robots, aerial or otherwise, is naturally abhorrent to us because robots--at least the robots that we have now--are incapable of making common sense decisions or distinguishing--with 100 percent accuracy--between friend or foe, surrendering troops or hostile enemy, the benign and the threatening.

But that capability gap between human and machine, as WaPo reports, is shrinking. The question is: when will it have shrunk enough that we trust robots with life and death decisions? As we’ve been coldly reminded by incidents in Iraq and Afghanistan, even highly trained soldiers don’t always make the right decisions on the ground. At what percentage of error are we willing to say autonomous robots are ready for war?

Click through below for the Post piece. It’s a quick and engaging morning read.

[Washington Post]

“Those electrons feel GOOD!,” R2 tweeted yesterday as its visual systems were powered on. “One small step for man, one giant leap for tinman kind.”

Har har. But humanoid humor aside, this is a big step for those who have been following R2’s progress from the labs at NASA to his launch aboard Discovery to his arrival at the ISS (R2 has been tweeting all of this along the way). It is the first humanoid robot ever taken into space, and he very well could be the model for many future generations of humanoid helper ‘bots launched aboard orbiting spacecraft and perhaps even on a future deep space mission. Along the way, future versions of R2 may even assist ISS astronauts during spacewalks.

Right now, R2 is only a waist-up humanoid; his torso is anchored to a pedestal from which he can use his arms to help human crew members in the orbiting lab. But a pair of legs is being designed for R2 and could launch to the ISS in 2013, at which point it will become something of a real life C-3PO capable of following around its human counterparts, getting into all kinds of scrapes, and providing somewhat flat comic relief.

[io9]

Troops require more rugged, easily deployable recon robots, and throwable bots fill the bill at this month's AUVSI drone convention

I’m at AUVSI’s (that’s the Association for Unmanned Vehicle Systems International) Unmanned Systems North America convention at the Recon Robotics booth, checking out the company’s Scout reconnaissance robot, a tiny two-wheeled system weighing just more than a pound. It’s a diminutive machine, about the size of a tallboy beer can. And I’ve just been invited to chuck it over an eight-foot wall.

At an exhibition boasting some seriously mean-looking machines (QinetiQ’s MAARS robot sports a 7.62-millimeter machine gun and four 40-millimeter grenade launcher tubes) these little throwbots (that's a throwable robot, and also a brand name of one of Recon's variants) aren’t exactly an intimidating sight. But their portability and durability serve a mission profile that is becoming increasingly attractive to troops serving in theater, and at AUVSI’s robot roundup last week it showed. It seems like every company with a stake in the ground-based robotic recon game is bringing out a robot that you can literally hurl into action.

Tactical robots have gained a ton of traction within the military over the past decade for several reasons, not least of which is the fact that any time you send a robot to execute a dangerous task, you’re not sending a human. But robots are generally complicated, fickle machines packing a lot of moving parts (and a lot of extra weight). They often require soldiers to undergo special training just to learn how to use them.

What soldiers in, say, Afghanistan really need is a robot intuitive enough that any soldier can pick up the controls, and something highly portable and easily deployable on a moment’s notice so it can quickly begin feeding real-time intelligence to mission commanders when threats emerge on short notice.

In short, they need a small, simple robot they can throw or launch over walls or into second-story windows, something that soldiers can control with handheld Xbox-like controls that come naturally to young men and women of a certain generation, something that can beam them back video and/or other information from a safe distance (and as stealthily as possible). And at AUVSI that need was being met with a smattering of little robots that company reps couldn’t wait to let us abuse.

In a demo at AUVSI’s makeshift proving ground, I watched QinetiQ’s Dragon Runner series ‘bots roll up and over obstacles and dig in the sand for the lead wires of mock IEDs. The newest addition to the QinetiQ family, the Dragon Runner 10, is an 11-pound variant of its larger brother, the 20-pound Dragon Runner 20 which has been on the market for some time now.

Both are man-portable, but the biggest difference between Dragon Runner 10 and Dragon Runner 20 (besides the obvious weight/portability difference): Dragon Runner 10 is being billed as throwable at ranges up to 12 feet. That’s no 100-plus-foot plunge from a helicopter, but it’s enough to make it over a high wall when Marines want to see what’s waiting on the other side--in IR-enhanced low-light vision if necessary.

And Dragon Runner 10 has some other serious merits. Though heavy at more than 10 pounds, it has a 5-pound payload capacity that can accommodate extras like a robotic arm (which is how it was able to dig in the aforementioned sand for IEDs). It’s a throwable ‘bot that does more than just reconnaissance--at the cost of a little added weight.

Roll out with iRobot’s FirstLook and you can cut that poundage by more than half. FirstLook is purely a recon ‘bot--no manipulator robot arm here--but at just 5 pounds it offers the same IR capability and has front, rear, and side-facing cameras. It’s a two-tracked robot resembling a seriously pared down version of iRobot’s renowned PackBot. And the wrist-mounted touchscreen display lends that air of warfighter-of-the-future we’ve come to expect from iRobot.

Throwability is pretty good too, as iRobot claims FirstLook can manage drops onto concrete from 15 feet. If the company’s representatives have any misgivings about this claim, it doesn’t show. They were happy to toss FirstLook carelessly up and down the aisles at AUVSI purely for my--and their--amusement. And the robots never missed a step, save the few seconds it took for them to right themselves when they landed upside-down.

But for pure throwability, there’s still nothing like Recon Robotics’ Scout. The whole system is practically featherlight at 1.3-pounds (plus a couple more for the control unit, for a total of 3.2 pounds). Both robot and control unit can charge from the battery systems troops already carry with them into the field--no added weight in power supplies--and its titanium shell and cast urethane wheels ensure a listed drop shock resistance of 30 feet--twice that of any other we saw at AUVSI.

Its throw shock resistance is listed at four times that distance (that’s a 120-foot radius from the operator where this Throwbot can be deployed via hail mary or a launcher device Recon has in the works). It’s capable of the same IR night vision as Dragon Runner and iRobot, but is hands-down the smallest, most discreet of the three.

Needless to say, I didn’t have the space (or, likely, the arm) to test the limits of the Scout’s range at AUVSI. But I did get to throw it onto the roof of a faux building the company had constructed and then, with it out of view, drive the thing around via the tiny video screen embedded in the single-joystick controller that’s so intuitive and easy to use that I felt like a pro after ten seconds at the stick.

Hurling a robot overhand and then driving it around remotely is just as much fun as it sounds, but it also makes it demonstrably clear just how easy and effective these kinds of robots could be if, say, you're about to commit several breathing humans to breaching a walled compound.

And I'm not the only one who thinks so. Dragon Runner and FirstLook are brand new, but Recon's first variants of the current line of Throwbots have been available for a few years. To date, they have put 2,000 of them into service. Virtually every agency with reason to worry about what's around the corner has deployed them: the U.S. military, various police and firefighting departments, the FBI, DEA, U.S. Marshals, the State Department (for embassy security), the Border Patrol, and more than 200 SWAT teams. Just today, U.S. Special Operations Command announced that it is buying nearly 400 more.

QinetiQ's and iRobot's models will likely find similar traction alongside other small, throwable 'bots that are likely incubating in labs elsewhere, because they meet a critical need for fast, actionable situational awareness on the battlefield. It's been said that knowing is half the battle. It looks as if throwing could become an integral part of it as well.

Unveiled at this week's giant drone conference

When mines collapse, the biggest hindrance to a speedy search and rescue operation is the lack of information. Mining accidents generally bring about a buffet of dangerous conditions: structural weaknesses within the shafts themselves, poisonous vapors, explosive gases, flooded tunnels, etc. Rescue crews can’t charge into such conditions without proper reconnaissance, lest they risk compounding the situation by creating a second disaster on top of the first.

Gemini-Scout is designed to cope with all of these things so it can get down into a mine quickly, searching for survivors and assessing threats so human searchers can get into place as quickly as possible. Its tracked propulsion and articulated suspension allow it to climb rubble piles and crawl over uneven terrain. In the ground demo area at AUVSI, Gemini rolled easily over stair-like obstacles and into the sand and gravel pits, turning tight circles and kicking up a mess before climbing out just as easily. It took 45-degree climbs with no serious problems, and at less than two feet tall it maneuvered through tight spaces with ease.

But more specifically to its purpose, Sandia engineers explained, Gemini-Scout can move through up to 18 inches of water while sampling the air for toxic fumes (technically it can operate through deeper water, but doing so would immerse the air sensors on its mast). Those air measurements are critical because the data they collect paves the way for manned rescue operations. They also let rescue personnel know if they are dealing with explosive methane gases or other flammable vapors.

To that end, Gemini-Scout’s electronics are packed in explosion-proof casings. A blast triggered by something else might disable the robot, but its own electronics won’t provide sparks that could trigger a second explosion and complicate a search and rescue operation.

A thermal camera helps Gemini-Scout search for survivors and two way radios allow handlers to communicate with any survivors the robot locates. The ‘bot can even be configured to carry food, air tanks, or other supplies to trapped miners, or to drag them to safety.

And because mine disasters can happen unexpectedly anywhere in the world, Sandia engineers wanted to make it operable by just about anyone. Gemini-Scout is controlled with a standard Xbox 360 remote, so virtually anyone comfortable with Call of Duty can answer the call of duty in a time of crisis.