Latest in Robotics news thread

DURUS: SRI's Ultra-Efficient Walking Humanoid Robot [30 times more efficient than ATLAS]
While disaster robots were making their way through the DARPA Robotics Challenge courses, over in the exhibit area outside, there was another competition taking place: an endurance challenge, also sponsored by DARPA, where robots from Sandia National Labs and SRI International slowly walked on treadmills with the goal of demonstrating how ultra-efficient they could be.
What does ultra-efficient mean in the context of walking robots? Think humanoid walking that’s 20 to 30 times more efficient than than Boston Dynamics’ ATLAS. A full size humanoid robot with that level of efficiency would able to operate for anywhere from 4 to 6 hours on a single charge.
The robot that SRI displayed at the DRC is called DURUS (the root there being “durable”), which was developed under a (super duper secret, apparently) DARPA contract beginning in 2013.
That year was also when Boston Dynamics unveiled their ATLAS robot, and that’s not a coincidence: DARPA saw ATLAS, immediately realized that power requirements made ATLAS an entirely impractical platform for real-world use, and started funding a program to develop ultra-efficient actuators. This was expanded into an efficient mobility challenge, with the goal of building a humanoid architecture that was similar to the ATLAS robot (with the same kinematics), except 20x to 30x more efficient.
 
"There are no human beings" Australian container port goes (almost) fully automated.

At Sydney's Port Botany, every hour of every day, the robots are dancing.

Well, they look like they are dancing – these 45 so-called AutoStrads, or automated straddles, machines that have taken on the work that until a couple of months ago was at least in part performed by dockworkers.

Almost 20 years ago, the Patrick container terminal at Botany played host to one of the most divisive industrial battles in Australian history, as the stevedoring company attempted to break the back of its union-dominated workforce.
In some respects that battle was won in April.

It was then that Patrick introduced, following a four-year investment program, a level of automation into its stevedoring operation that might be unsurpassed in the world.

"This is fully automated, there are no human beings, literally from the moment this truck driver stepped out of his cabin from then onwards this AutoStrad will take it right through the quay line without any humans interfacing at all," Alistair Field, the managing director of Patrick Terminals and Logistics, a division of Asciano, said on Wednesday.
 
Google DeepMind Teaches Artificial Intelligence Machines to Read

A revolution in artificial intelligence is currently sweeping through computer science. The technique is called deep learning and it’s affecting everything from facial and voice to fashion and economics.
But one area that has not yet benefitted is natural language processing—the ability to read a document and then answer questions about it. That’s partly because deep learning machines must first learn their trade from vast databases that are carefully annotated for the purpose. However, these simply do not exist in sufficient size to be useful.
Today, that changes thanks to the work of Karl Moritz Hermann at Google DeepMind in London and a few pals. These guys say the special way that the Daily Mail and CNN write online news articles allows them to be used in this way. And the sheer volume of articles available online creates for the first time, a database that computers can use to learn and then answer related about. In other words, DeepMind is using Daily Mail and CNN articles to teach computers to read.
The deep learning revolution has come about largely because of two breakthroughs. The first is related to neural networks, where computer scientists have developed new techniques to train networks with many layers, a task that has been tricky because of the number of parameters that must be fine-tuned. The new techniques essentially produce “ready-made” nets that are ready to learn.
 
AI program predicts key disease-associated genetic mutations for hundreds of complex diseases
A decade of work at Johns Hopkins has yielded a computer program that predicts, with far more accuracy than current methods, which mutations are likely to have the largest effect on the activity of the “dimmer switches” (which alter the cell’s gene activity) in DNA — suggesting new targets for diagnosis and treatment of many diseases.

A summary of the research was published online today (June 15) in the journal Nature Genetics.
“Our computer program can comb through the genetic information from a specific cell type and predict which ‘dimmer switch’ mutations are most likely to alter the cell’s gene activity, and therefore its function,” says Michael Beer, Ph.D., associate professor of biomedical engineering at the Johns Hopkins University School of Medicine.
 
Softbank, Foxconn and Alibaba are joining to sell $1600 humanoid robots like Pepper

pepperbot.jpeg

Foxconn Technology Group, Japan’s Softbank Corp and Alibaba are planning to set up a joint venture to produce robots, Japan’s Nihon Keizai Shimbun reported yesterday. Foxconn, also known as Hon Hai Group, and Softbank have agreed on the joint venture deal, which is scheduled to launch production by the end of this year, with an annual capacity
...
 
No Strings on the running and hurdling Cheetahbot

MIT-Jumping-Cheetah-1.jpg

MIT researchers who built a robotic cheetah have now trained it to see and jump over hurdles as it runs — making this the first four-legged robot to run and jump over obstacles autonomously and untethered. The robot can “see,” with the use of onboard LIDAR — a visual system that uses reflections from a laser to map terrain. The team developed...
 
TIL: A scientist let a computer program a chip, using natural selection. The outcome was an extremely efficient chip, the inner workings of which were impossible to understand.

For the first hundred generations or so, there were few indications that the circuit-spawn were any improvement over their random-blob ancestors. But soon the chip began to show some encouraging twitches. By generation #220 the FPGA was essentially mimicking the input it received, a reaction which was a far cry from the desired result but evidence of progress nonetheless. The chip’s performance improved in minuscule increments as the non-stop electronic orgy produced a parade of increasingly competent offspring. Around generation #650, the chip had developed some sensitivity to the 1kHz waveform, and by generation #1,400 its success rate in identifying either tone had increased to more than 50%.
Finally, after just over 4,000 generations, test system settled upon the best program. When Dr. Thompson played the 1kHz tone, the microchip unfailingly reacted by decreasing its power output to zero volts. When he played the 10kHz tone, the output jumped up to five volts. He pushed the chip even farther by requiring it to react to vocal “stop” and “go” commands, a task it met with a few hundred more generations of evolution. As predicted, the principle of natural selection could successfully produce specialized circuits using a fraction of the resources a human would have required. And no one had the foggiest notion how it worked.
 
Scientist develops model for robots with bacteria-controlled brains


Forget the Vulcan mind-meld of the Star Trek generation -- as far as mind control techniques go, bacteria is the next frontier.
In a paper published July 16 in Scientific Reports, which is part of the Nature Publishing Group, a Virginia Tech scientist used a mathematical model to demonstrate that bacteria can control the behavior of an inanimate device like a robot.
"Basically we were trying to find out from the mathematical model if we could build a living microbiome on a nonliving host and control the host through the microbiome," said Ruder, an assistant professor of biological systems engineering in both the College of Agriculture and Life sciences and the College of Engineering.
"We found that robots may indeed be able to have a working brain," he said.
Ruder spoke about his development in a recent video:
 
Hitchhiking robot embarking on coast-to-coast tour across US


With its thumb raised skyward and a grin on its digital face, the robotic creation of two researchers in Canada is about to start a hitchhiking journey across the U.S.

The humanoid robot named hitchBOT has already caught rides across Canada and in Europe, relying on the kindness and curiosity of strangers. But on Friday it starts its first U.S. tour, setting out from Salem, Massachusetts, with dreams of San Francisco ahead.

Along the way, it hopes to see some quintessential American sites, including Times Square, Mount Rushmore and the Grand Canyon.

The kid-size robot is immobile on its own, so it gets from place to place by being schlepped around by strangers. Travelers can pass it off to others they meet, or leave it at a gas station or shop. They can just leave it seated on its kickstand with its thumb raised.

Ideally, the creators hope, drivers won't leave the bot along busy roads and will charge the battery when it runs low. Otherwise, there are no rules.

"We want to see what people do with this kind of technology when we leave it up to them," said Frauke Zeller, one of the creators and an assistant professor in professional communication at Toronto's Ryerson University. "It's an art project in the wild—it invites people to participate."
 
Algorithms Based on Brains Make For Better Networks


When it comes to developing efficient, robust networks, the brain may often know best.

Researchers from Carnegie Mellon University and the Salk Institute for Biological Studies have, for the first time, determined the rate at which the developing brain eliminates unneeded connections between neurons during early childhood.

Though engineers use a dramatically different approach to build distributed networks of computers and sensors, the research team of computer scientists discovered that their newfound insights could be used to improve the robustness and efficiency of distributed computational networks. The findings, published in PLOS Computational Biology, are the latest in a series of studies being conducted in Carnegie Mellon’s Systems Biology Group to develop computational tools for understanding complex biological systems while applying those insights to improve computer algorithms.
 
Japanese consumers quick to welcome robot assistants

Once the preserve of automated factories and research and development laboratories, androids are stepping out into Japanese society. DW's Tokyo correspondent Julian Ryall reports.
In September, 11 new members of staff are scheduled to start work at Tokyo's Haneda International Airport. The latest additions to the workforce will be tasked with menial jobs, such as cleaning the concourse and transporting luggage, but they are unlikely to complain about their responsibilities as they are robots.
The operator of the airport has signed an agreement with Cyberdyne Inc. to put the technology company's latest creations to the test in a real-life work environment. And, should the experiment be deemed a success, then both parties say they are ready to deploy more non-human members of staff in the future.
Putting androids to work at Haneda might be seen as something of a gimmick or a publicity stunt, but there is a more serious reason behind the introduction of robots at the airport - as well as in an increasing number of service sector positions across Japan.
 
GE atomic swimmer robot keeps tabs on nuclear reactors
By David Szondy - August 3, 2015 2 Pictures

One truism of nuclear reactors is that you really don't want to be next to one. Unfortunately, reactor cores need to be inspected and maintained, which means teams of workers going inside the containment vessel. It's an operation that's not only hazardous, but expensive and time consuming. In an effort to make such inspections safer, cheaper, and faster, GE Hitachi Nuclear Energy has developed the Stinger; a free-swimming, remote-controlled robot that replaces humans for cleaning and inspecting reactor vessels.
 
Automation, Algorithms, And Android News Readers: How Robots Are Changing The Face Of Journalism
The year is 1986, and radio listeners in Adelaide are shocked. ABC Local Radio has just announced that its newsreaders are being replaced by advanced voice synthesis robots. The station blames reports that even the tone of someone’s voice could introduce bias to a neutrally written story. Veteran radio newsreader Alf Jarvis, fresh from announcing his retirement, rails against this change and declares he is glad to be getting out of the industry. A spokesman for the public sector union also voices their concern for job losses.
But the date is April 1.
Technology writer and broadcaster Stilgherrian fondly recalls his prank: “I was one of the producers working on the 8:30am daily talk show. That morning we ran the story straight after AM, and the talkback board lit up with outraged callers. We even used one of the early Mac computer programs to read the day’s news. It was great fun.”
Fast forward thirty years, and robot journalists are closer to becoming a reality. In June this year, Osaka University Professor Hiroshi Ishiguro unveiled Kodomoroid: the world’s first news-reading android. After years of research and development it can pronounce complex tongue twisters, speak multiple languages and interact with people. It looks like this...
 
Roomba just got government approval to make an autonomous lawn mower
It's been nearly a decade since the earliest whispers suggested iRobot, makers of the Roomba, were building a lawn mower. But we seem to be a bit closer to the future we were promised: the FCC has granted approval to iRobot to build a hands-free mowing-bot, Reuters reports.
Although we don't know all of the specifics, the mower, according to Reuters, would operate through stakes in the ground that wirelessly connect to a mower and map out where it should cut. That approach required a waiver from the Commission, which was granted despite objections from the National Radio Astronomy Observatory. The observatory argued the mower's signal would interfere with telescopes, but the FCC sided with iRobot, saying its limitations would insure astronomers' work wasn't harmed.
But a mower still doesn't sound like it will be available to consumers imminently. According to Reuters, iRobot says the waiver will let it "continue exploring the viability of wideband, alongside other technologies, as part of a long-term product exploration effort in the lawn mowing category."
 

Forum List

Back
Top