Robotica Exotica

Autonomous Speedboat

This is interesting. Especially since I was just talking to some people about building one to help with environmental monitoring. A company called MRVI is marketing an autonomous speedboat for patrol operations. Pretty cool looking, too.

MRV International, an innovative marine technology company, is providing state-of-the-art software and hardware for robotic USVs (unmanned surface vessels) to man the US's harbours and waterways.

Operating as autonomous marine robotic platforms, the USVs use MRV's new-generation software for remote command and control, navigation, route planning, event and crisis management, onboard diagnostics and other functions. Linked with broadband wireless telemetry, the USVs can be operated from virtually anywhere in the world via the Internet.

To identify terrorism threats, the USVs can be equipped with radar, sonar, and video surveillance capabilities, as well as on-site, automated chemical, biological and radiological analysis. The all-weather, day-and-night robot boats also have application in piracy, narcotics interdiction, illegal immigration, chemical and radiological spill detection and many other areas.

ROBOlymics

San Francisco will be hosting the first ever ROBOlympics competition. This should be pretty neat, though most of the entries appear to be remote controlled rather than trully autonomous robots. Still, there should be a lot of good stuff to look at.

Fearsome BattleBots, fearless fire-fighting robots, finely-tuned soccer-playing robot teams, sumo bots and robots built by students will be on display at the SF Commonwealth Club in a rock-em, sock-em preview of the first ever international Robot Olympics—ROBOlympics—to be held in San Francisco next month.

Robot-builders young and old, American, English, and Japanese, will be in attendance to provide live demonstrations and explanations of their ROBOlympians. This will be an exclusive chance to get close to the builders—amateur engineers from many walks of life with dramatic stories—before they enter into the heat of competition. Interviews may be conducted in a friendly atmosphere with complimentary cocktails and hors d’oeuvres.

While many different robot events exist around the world, ROBOlympics will be the first ever to bring them all together in a single place and time. San Francisco is the birthplace to such world famous robot competitions as the internationally televised BattleBots and Robot Wars, and ROBOlympics represents a triumphant return of this booming mechanical pastime to its rightful home—while including robots competitions from around the world, both long-standing and nascent.

DARPA Grand Challenge Ends with a Whimper

Well, as I had predicted earlier, no vehicle finished the Grand Challenge. In fact, the one that went the farthest, went less than 8 miles. Which, by the way, is still a tremendously impressive feat. This just goes to illustrate the difficulty of doing autonomous navigation.

The Grand Challenge event was supposed to be a 10-hour sprint across the desert, with a $1 million prize to the designers of the first driverless vehicle to transit 142 miles of sand and rock from Barstow (San Bernardino County) to Primm, Nev., just across the state line.

But shortly after 11 a.m., Anthony Tether, director of the Defense Advanced Research Projects Agency, the Pentagon office that put up the prize, took the stage at a casino near the supposed finish line to announce: "The Grand Challenge ended about 10 minutes ago when the last 'bot went out.''

Despite the race's somewhat comic end -- the robotic dirt bike entered by an Albany man toppled two feet from the starting gate -- Tether said the competition had advanced DARPA's aim of spurring the development of driverless combat vehicles capable of fighting desert wars without putting soldiers in harm's way.

"It exceeded our expectation by the amount of people who showed up, and by the types of people who showed up,'' said Tether, saying the agency would probably stage a new challenge in a year or so after the volunteers have time "to get their batteries recharged.''

DARPA sponsored the Grand Challenge because a decade of government-funded contracts with defense firms has failed to produce breakthroughs. The agency hoped the prize and prestige of the challenge would entice academic and garage inventors to come up with new ideas. Tether said DARPA spent about $13 million to sponsor the event, and got more than its money's worth in promising new approaches.

Flexibot Robotic Arm for Helping the Disabled

The Flexibot is a pretty neat concept in robotic assistance for the disabled. The original idea was a robotic arm that was attached to a wheel-chair, but they have gone one step farther and allowed the arm to now detach from the wheelchair and move end over end through fixed attachment points. It is a pretty neat idea, but without any sensors other than internal angles, I'm not sure how accurate it can be.

Unlike most other mobile robots, Flexibot gets about by flipping end over end from one docking station to the next. The docking points serve the dual purpose of both supporting and powering the droid.

Each arm weighs about 11 kilograms and its five motors making it capable of carrying up to four kilograms. The docking station uses a simple bayonet fitting, much like a light bulb, says Gunnar Bolmsjý, a mechanical engineer at Lund University, in Sweden, who designed it.

This mechanism also allows the robot to attach household devices, such as electric toothbrushes, to either of its ends. In April, the team will add fold-out robotic grippers, which will mean the arm can grasp any object, not just those specially adapted for use.

At the moment the robot has no sensors and navigates simply by logging how far it has already moved. This approach allows an accuracy of one tenth of a millimetre, say the team. Users control the arm either by blowing down a straw or pressing a single button.

The Robots are Coming

The robots are coming! And when they get here, they'll take out the trash. I just love that. CNET has an overview of the different types of robots out there, and what they can do and where they want to go. Overall, it is a good primer on the state of the art.

The idea of automatons that can perform various tasks has been around since ancient Egypt. The word "robot," however, is of relatively recent vintage, coined by Czech playwright Karel Capek in the 1921 play "R.U.R."

The first commercial robots appeared in the early 1960s as the world was consumed with the Atomic Age science of the Cold War. Unimation, founded by Engelberger, created robotic manufacturing arms, while Barrett Electronics came out with a driverless electric cart for grocery warehouses that was navigated by signal-emitting wires in the floor.

It wasn't an easy sell. "I had to go to 46 different companies, including GM and IBM, before I got some money from a railroad company," Engelberger recalled. Although GM passed the first time, it eventually became the first company to install a Unimation machine.

Asimo's Creator Speaks

The creator of Honda's Asimo robot, Edgar Korner, speaks about the future of humanoid robots in this interview in e4engineering.com. This is an interesting peek into the mind of a futurist, and I think that he comes off well.

But even if the robot is safe, fun and useful, is there anything to suggest that people will want it? Korner thinks so, and claimed that the market for domestic androids will be 'really huge.' He pointed to the trend in Japan for robot pets, which are no longer marketed simply as expensive toys but are also sold to pensioners as both companions and protectors. Able to respond to basic stimuli, these machines will even alert the emergency services if their ageing owners go quiet on them.

And then there is the inevitable question for every engineer working in the robotics field: what about the 'Terminator scenario' - the Hollywood-fuelled fear that robots will eventually become so intelligent that they enslave the human race?

Korner's response is as unequivocal as you would expect from someone devoted to this area of technology. 'Any tool made by humans to make life easier can potentially be turned into a weapon. Should we stop making knives? There are easier ways to kill people than to develop autonomous robots, and I don't think it's a reasonable fear that humanoid robots will dominate and control the world.

'I expect robot development to set humans free to indulge our creative arts and our capabilities to deal with creative processes. By overcoming the limitations set by the limited capacity of our brain for storing information, we will be able to develop new areas of our creative designs and develop systems that were previously impossible.'

Grand Challenge Gearing Up

The NY Times is publishing some more information on the DARPA Grand Challenge, highlighting the difficulties as well as the impressive work that has been generates. I am delighted to see the buzz being generated by this, as I think it is a good contest, with minimal ways in which to cheat.

The Pentagon, under a mandate from Congress to save lives by turning to unmanned combat vehicles to meet a third of its needs by 2015, has become impatient with its usual crowd of big-name military contractors, like the Lockheed Martin Corporation and the General Dynamics Corporation, to come up with a solution. It turned instead to the spur of free market capitalism, inspiring a motley band of computer scientists, artificial intelligence experts and robot lovers to take on the challenge.

It is not clear whether any of the couple of dozen vehicles expected to line up at the start will be able to complete the task. After first passing a preliminary test scheduled to begin today, the winning machine will then have to navigate unaided at an average speed of about 20 miles an hour through a desert strewn with boulders, trees, brush, potholes and possibly the odd porcupine or donkey.

NYTimes on Robotic Vacume Cleaners

The NY Times has an article in their circuits section on the various robotic vacume cleaners. I have to admit that I own a Roomba, and I like it. I have also dismantled the Trilobyte, which is an interesting robot in its own right. You do get used to having clean floors without much fuss after owning one.

While the idea of a vacuum that cleans while its owner does almost anything else may sound appealing, the robotic models are in many ways like early dishwashers and washing machines, said Carolyn Forte, the director of home care at the Good Housekeeping Institute in New York. That is, initially, they don't measure up to what they replaced.

"People who washed dishes and clothes by hand probably didn't think dishwashers and washing machines worked very well when they were introduced," she said. "But now we can't live without them."

For now, Ms. Forte said, the robotic vacuums are best at picking up surface dirt and dust, and - because of their slick disk shape - at reaching places that some vacuums can't get to, like under the sofa. But the robotic sweepers don't do stairs or cushions, they are not as powerful as conventional vacuums, and they sometimes miss spots. So for deep cleaning, Ms. Forte said, "you unfortunately still have to drag out the old vacuum cleaner."

Berkeley develops an Exoskeleton

This is pretty neat. Berkeley has developed a robotic exoskeleton that straps onto the legs of a human pilot, and adds power and lifts loads. This has been a staple of SciFi stories for a while, but it is interesting to see it experimentally demonstrated.

The researchers point out that the human pilot does not need a joystick, button or special keyboard to "drive" the device. Rather, the machine is designed so that the pilot becomes an integral part of the exoskeleton, thus requiring no special training to use it. In the UC Berkeley experiments, the human pilot moved about a room wearing the 100-pound exoskeleton and a 70-pound backpack while feeling as if he were lugging a mere 5 pounds.

The project, funded by the Defense Advanced Research Projects Agency, or DARPA, began in earnest in 2000. Next week, from March 9 through 11, Kazerooni and his research team will showcase their project at the DARPA Technical Symposium in Anaheim, Calif.

For the current model, the user steps into a pair of modified Army boots that are then attached to the exoskeleton. A pair of metal legs frames the outside of a person’s legs to facilitate ease of movement. The wearer then dons the exoskeleton’s vest that is attached to the backpack frame and engine. If the machine runs out of fuel, the exoskeleton legs can be easily removed so that the device converts to a large backpack.

More than 40 sensors and hydraulic actuators form a local area network (LAN) for the exoskeleton and function much like a human nervous system. The sensors, including some that are embedded within the shoe pads, are constantly providing the central computer brain information so that it can adjust the load based upon what the human is doing. When it is turned on, the exoskeleton is constantly calculating what it needs to do to distribute the weight so little to no load is imposed on the wearer.

Japanese Two-legged Home Robot

Slashdot has a link to an article on a Japanese athletics company that is going to be selling a two-legged robot for home use. Someone has translated the article from Japanese:

A two legged robot that you can play with in your home will be available as early as the end of this year for 500000 yen. Venture company ZMP and sports equipment company Mizuno, together announced on the 2nd.

As for 2 legged robots, Sony's QRIO and Honda's ASIMO are famous, but they are mainly for publicity purposes and are rarely purchased for general use.

nuvo is 39 centimeters, and 2.5 kilograms. It can walk forward, backward, left, or right, and if it falls, can get up automatically from any position. It can be taught about 1000 words, follow directions, (something [jishiki?]) or dance.

In the head is a camera. Using NTT DoCoMo 3rd generation phones, you can see from the robot's perspective, and use it as a remote controlled "security robot".

Minimizing the use of joints allowed the cost to remain low. Over 3000 orders are expected by the new year.

Very, very neat. I don't think it will go much beyond the toy-like aibo, but still pretty cool.

Robo-Snail

Researchers at MIT have created a robot that imitates the locomotion of a snail. At first blush, this might be considered to be against the direction that we generally want robots to go, but snails have a very robust motion system, able to climb over obstacles that would render a wheeled vehicle useless.

Real snails travel on their single foot along a trail of mucus, a slimy fluid secreted by a special gland in the foot. Snails move by pushing the mucus between their foot and the ground or other surface. Snail mucus is an example of a non-Newtonian fluid, a type of fluid of particular interest at MIT’s Hatsopoulos Microfluidics Laboratory, where Hosoi and Chan work. Unlike Newtonian fluids such as oil or water, non-Newtonian fluids change their properties and thicken when they're subjected to stress, such as downward pressure from a snail's foot. Since snails use different mechanisms to move, Hosoi and Chan designed two RoboSnails, each mimicking a different method of snail locomotion.

The team's first RoboSnail imitates a snail mechanism that Hosoi compares to the up-and-down motion of ocean waves. "There's a wave that propagates along the bottom of the snail’s foot," explains Hosoi. "And the wave can either propagate in the direction that the snail is moving, or it can propagate backwards." The first RoboSnail's version of mucus is silicon oil combined with clay particles. Its body is a series of plastic plates strung along a wire helix that is connected to a small motor Chan borrowed from a toy car. When the motor is running, the plates create a wave in a rubber foot.

Primer on Controller Area Network

I just stumbled across this good summary of the Controller Area Network (CAN) bus over at Texas Instruments. If you are interested in using this bus on one of your systems, this is a very good overview.

A controller area network (CAN) is ideally suited to the many high-level industrial protocols embracing CAN and ISO 11898 as their physical layer. Its cost, performance, and upgradeability provide for tremendous flexibility in system design. This paper presents a brief introduction to the CAN operating principles, the implementation of a basic CAN bus using Texas Instrument’s CAN transceivers and DSPs, and a discussion of the robust error detection and fault confinement mechanisms. Some of the properties of CAN, especially relating to the electrical layer and features of transceiver products, are then discussed at a tutorial level.