UPDATE (1/25/2018): NASA has successfully unfrozen Curiosity, which will now live to rove another day.

But here's the original report shared by a reader detailing what the concerns were: NASA reports that Curiosity has suffered a system failure that left the robot unaware of its position and attitude on the red planet. Until it recovers, Curiosity is frozen in place. Mars is far enough away that we can't directly control Curiosity in real-time -- the rover gets batches of commands and then carries them out. That means it needs to have precise awareness of the state of all its joints, as well as environmental details like the location of nearby obstacles and the slope of the ground. This vital information ensures the rover doesn't bump anything with its arm or clip large rocks as it rolls along.

Curiosity stores all this attitude data in memory, but something went wrong during operations several days ago. As the rover was carrying out its orders, it suddenly lost track of its orientation. The attitude data didn't add up, so Curiosity froze in place to avoid damaging itself. While the rover is physically stuck in place, it's still in communication with the team here on Earth. Since everything else is working on the rover, NASA was able to develop a set of instructions that should get the rover moving again. When transmitted, the data will inform Curiosity of its attitude and confirm its current state. This should allow the rover to recover and keep performing its safety checks. However, NASA also hopes to gather data on what caused the issue in the first place. The hope is they can avoid another freeze-up in the future.

nickwinlund77 shares a report from Corvallis Gazette-Times: A petition to limit each grocery store to two self-checkout kiosks can move forward to signature gathering for a state ballot measure. On Friday, the Oregon Supreme Court certified the attorney general's description of the proposed measure. Backers need 112,020 signatures to get to voters' ballots in November. Filed in July, Initiative Petition 41 is backed by the Oregon AFL-CIO, a coalition of labor groups representing about 300,000 Oregon workers.

"We have been consistently concerned about the impacts of technology and automation on the livelihoods of working people, especially when they have no voice in how technology is used in their workplaces," Graham Trainor, president of the Oregon AFL-CIO, said in a statement. "You can see expansion of self-checkout machines in stores across the country and in Oregon." He said jobs are lost as a result. The AFL-CIO contends self-checkout kiosks make customers feel socially isolated, particularly elderly people, and that the kiosks let stores rely more on part-time workers and leaves workers "feeling devalued." They also claim self-checkout stands make it easier for minors to buy alcohol and for people to steal from stores. The measure would give the state Bureau of Labor and Industries enforcement power and let it issue penalties for stores that provide too many self-service stations. "Today's customer wants convenience and less hassle when shopping," said Joe Gilliam, president of the Northwest Grocery Association, an industry group. "This is evident in the growth of online shopping for local pick-up and home delivery. This measure is tone deaf to what the public is demanding in the marketplace."

He said that self-checkout lets customers check out more quickly and privately. He said presuming that self-checkout machines would replace workers is "simply untrue."

A team at Facebook AI has created a reinforcement learning algorithm that lets a robot find its way in an unfamiliar environment without using a map. MIT Technology Review reports: Using just a depth-sensing camera, GPS, and compass data, the algorithm gets a robot to its goal 99.9% of the time along a route that is very close to the shortest possible path, which means no wrong turns, no backtracking, and no exploration. This is a big improvement over previous best efforts. [...] Facebook trained bots for three days inside AI Habitat, a photorealistic virtual mock-up of the interior of a building, with rooms and corridors and furniture. In that time they took 2.5 billion steps -- the equivalent of 80 years of human experience. Others have taken a month or more to train bots in a similar task, but Facebook massively sped things up by culling the slowest bots from the pool so that faster ones did not have to wait at the finish line each round.

As ever, the team doesn't know exactly how the AI learned to navigate, but a best guess is that it picked up on patterns in the interior structure of the human-designed environments. Facebook is now testing its algorithm in real physical spaces using a LoCoBot robot.

Researchers from the University Hospital Zurich, ETH Zurich, Wyss Zurich and the University of Zurich have developed a machine that repairs injured human livers and keeps them alive outside the body for one week. ScienceDaily reports: Until now, livers could be stored safely outside the body for only a few hours. With the novel perfusion technology, livers -- and even injured livers -- can now be kept alive outside of the body for an entire week. This is a major breakthrough in transplantation medicine, which may increase the number of available organs for transplantation and save many lives of patients suffering from severe liver disease or a variety of cancers. Injured cadaveric livers, initially not suitable for use in transplantation, may regain full function while perfused in the new machine for several days. The basis for this technology is a complex perfusion system, mimicking most core body functions close to physiology. The corresponding study was published in the journal Nature Biotechnology.

The inaugural study shows that six of ten perfused poor-quality human livers, declined for transplantation by all centers in Europe, recovered to full function within one week of perfusion on the machine. The next step will be to use these organs for transplantation. The proposed technology opens a large avenue for many applications offering a new life for many patients with end stage liver disease or cancer.

Cy Guy writes: Having not learned the lessons of Jurassic Park and the Terminator, scientists from the University of Vermont and Tufts have created "reconfigurable organisms" using stem cells from frogs. But don't worry, the research was funded by the Department of Defense, so I'm sure nothing could possibly go wrong this time. "The robots, which are less than 1mm long, are designed by an 'evolutionary algorithm' that runs on a supercomputer," reports The Guardian. "The program starts by generating random 3D configurations of 500 to 1,000 skin and heart cells. Each design is then tested in a virtual environment, to see, for example, how far it moves when the heart cells are set beating. The best performers are used to spawn more designs, which themselves are then put through their paces."

"Because heart cells spontaneously contract and relax, they behave like miniature engines that drive the robots along until their energy reserves run out," the report adds. "The cells have enough fuel inside them for the robots to survive for a week to 10 days before keeling over."

The findings have been published in the Proceedings of the National Academy of Sciences.

The last decade was a big one for artificial intelligence but researchers in the field believe that the industry is about to enter a new phase . From a report: Hype surrounding AI has peaked and troughed over the years as the abilities of the technology get overestimated and then re-evaluated. The peaks are known as AI summers, and the troughs AI winters. The 10s were arguably the hottest AI summer on record with tech giants repeatedly touting AI's abilities. AI pioneer Yoshua Bengio, sometimes called one of the "godfathers of AI", told the BBC that AI's abilities were somewhat overhyped in the 10s by certain companies with an interest in doing so. There are signs, however, that the hype might be about to start cooling off.

"I have the sense that AI is transitioning to a new phase," said Katja Hofmann, a principal researcher at Microsoft Research in Cambridge. Given the billions being invested in AI and the fact that there are likely to be more breakthroughs ahead, some researchers believe it would be wrong to call this new phase an AI winter. Robot Wars judge Noel Sharkey, who is also a professor of AI and robotics at Sheffield University, told the BBC that he likes the term "AI autumn" -- and several others agree.

A Chinese court has ruled that AI-generated works are entitled to copyright protection, in a win for tech giant Tencent. From a report: According to state media outlet China News Service (CNS), a court in Shenzhen this month ruled in favour of Tencent, which claimed that work created by its Dreamwriter robot had been copied by a local financial news company. The Shenzhen Nanshan District People's Court ruled that, in copying the Dreamwriter article, Shanghai Yingxun Technology Company had infringed Tencent's copyright. Dreamwriter is an automated writing system created by Tencent and based on the company's own algorithms. According to the reports, Shanghai Yingxun reposted a financial report written by Dreamwriter in August 2018 without Tencent's permission. The question of whether AI-generated works are protectable under copyright law have been the subject of much debate.

Walmart is testing back-of-store automated systems that can collect 800 products an hour, 10 times as many as a store worker. From a report: In the backroom of a Walmart store in Salem, N.H., is a floor-to-ceiling robotic system that the country's largest retailer hopes will help it sell more groceries online. Workers stand on platforms in front of screens assembling online orders of milk, cereal and toilet paper from the hulking automated system. Wheeled robots carrying small baskets move along metal tracks to collect those items. They are bagged for pickup later by shoppers or delivery to homes. Walmart is one of several grocers including Albertsons and Kroger that are using automation to improve efficiency in a fast-growing but costly business that comes with a range of logistical challenges.

The backroom robots could help Walmart cut labor costs and fill orders faster and more accurately. It also could address another problem: unclogging aisles that these days can get crowded with clerks picking products for online orders. A store worker can collect around 80 products from store shelves an hour, estimated John Lert, founder and chief executive of Alert Innovation, the startup that has worked with Walmart to design the system dubbed Alphabot. It is designed to collect 800 products an hour per workstation, operated by a single individual, Mr. Lert said. Workers stock the 24-foot-high machine each day with the products most often ordered online, including refrigerated and frozen foods. Fresh produce is still picked by hand in store aisles.

An anonymous reader quotes a report from CNBC: SoftBank-backed Zume is laying off 360 employees, accounting for about 50 percent of its workforce, and shuttering its robotic pizza business to focus on food packaging. SoftBank invested $375 million in Zume in 2018, giving the start-up a $1 billion valuation. Previously, Zume was valued at just $218 million and had risen $71 million in total, according to Pitchbook. Like other SoftBank-backed startups, Zume used the capital to quickly scale and increase its workforce. But, over the last year, investors have shifted their focus from "growth at all costs" to a clearer path to profitability.

Zume CEO and founder Alex Garden tells CNBC that it's a difficult day for the startup, but the changes being made will focus the business on "the inventions that are showing strong commercial traction." Garden says the company is creating 100 open roles in the Source Packaging unit that employees can reapply for. Pizza Hut has been testing Zume's round boxes on a limited basis. Zume's packaging -- which the company says is covered by a number of patents -- is made of sustainably harvested plant fiber and is industrially compostable. The robot pizza company, which launched in 2015, consisted of an army of robot sauce-spreaders and trucks packed full of ovens. Garden's goal at the time was to become the "Amazon of food."

The dogs, known as Aibos, are companion robots made by Sony -- robots that don't necessarily do much apart from providing company and comfort. From a report: Every Aibo -- Japanese for "companion" -- is manufactured identically, besides a choice between silver and white or a brown, black, and white version. They all have rounded snouts that include a camera for facial recognition capability, large, oval eyes to reveal their expressions, and a body that can turn on 22 different axis points to give them a range of motion. The owner decides the gender when they set them up, which determines the pitch of its bark and how it moves. They're cute. They know when you're smiling. And through machine learning and recognizing people with its camera, Aibos also shift their personality over time based on their interactions with people they spend time with. Soon, they become much more than a store-bought toy.

Still in the "off" position in the cafe, the Aibos' paws remained outstretched and their heads turned to one side. But one by one, as their owners kneeled down to turn them on from a switch at the scruff of their neck, each came to life. The screen of their doll-like eyes blinked open, they lifted their heads, stretched out their plastic limbs, and leaned back on their hind legs before standing on all fours. Almost like real dogs, they shook their heads as if to ward off sleep after a nap, wagged their tails, and barked.

An anonymous reader quotes CNN: A North Carolina couple was watching a movie in their bedroom when they suddenly heard loud noises coming from downstairs. Worried that it was an intruder, the two called 911.

The couple waited for police to arrive, hoping their 2-year-old daughter sleeping in her room wouldn't get up to check on the noise, said Thomas Milam, the husband, in his Facebook post shared to Forsyth County Sheriff's Office's page... Minutes after they called 911, police entered the home and began to search for an intruder. When the 911 operator told Milam to go downstairs to talk to the police, he said, the officers just had one question.

"Is this Roomba yours?"

Police had apprehended the suspect: the couple's brand new robotic vacuum. Milam said in the Facebook post that the vacuum had turned itself on in the night and gotten stuck in the hallway, where it had been repeatedly banging against the walls and making the sounds the Milams feared was an intruder.

An anonymous reader quotes a report from Los Angeles Times: There was only one job that robots could not do when Fast Retailing, the owner of Uniqlo, replaced 90% of its workers with robots at its flagship warehouse in Tokyo last year. But now, with the help of a Japanese start-up called Mujin, the world's third-largest retailer says it has cracked the final barrier to full automation, a priority for Uniqlo as Japan's aging population creates labor shortages. The two companies have invented a robot with two arms that can pick up soft T-shirts and place them neatly in boxes to be shipped to customers.

While it sounds easy, the ability to lift soft textiles has been a challenge for clumsy robotic arms. Add to this the need to sort through constantly changing seasonal clothes, in shades that are hard to distinguish and wrapped in various forms of packaging, and humans have always come out on top. The jointly developed robot, which was made by Yaskawa Electric Corp., is already operating in Fast Retailing's main warehouse in Tokyo, but Takino admitted that the robot was not able to handle all of the facility's products, and that it needed further development. For instance, the plastic packaging of the thermal underwear in Uniqlo's Heattech line is relatively simple for the robots to pick up, but this could become more difficult as Fast Retailing aims to switch to more eco-friendly paper bags. The robots are able to pick up belts, but they typically become unbundled as they are dropped into boxes. One solution would be for Fast Retailing to ensure that belts are sold in bundled forms. "We've been putting off working with an apparel company because it's so difficult," said Issei Takino, co-founder and chief executive of Mujin. "But Fast Retailing's strength is its ability to overhaul its entire supply chain to make it fit for automation. If we're going to take on this challenge, we had to do it with Fast Retailing."

Rumbling quakes on the red planet have been traced back to Cerberus Fossae, suggesting this geologically young region is still alive and cracking. From a report: Millions of miles away, a robot geologist stands alone on the dusty surface of Mars, listening for faint seismic echoes in the ground below. Its finger on the red planet's pulse is sensitive enough to pick up the whoosh of wind, the drone of dust devils, the creak of tectonic cracks, and many other rumbles ricocheting though the planet's insides. While most of these signals have been indistinct murmurs, two have stood out loud and clear, allowing scientists to trace them back to their source: the first active fault zone yet found on the red planet. Known as marsquakes, the events clocked in between magnitude 3 and 4, according to data from NASA's InSight lander presented at a recent American Geophysical Union conference. While the two quakes are small by Earth standards, they're among the largest yet detected on Mars. Scientists were able to trace both quakes to an area known as Cerberus Fossae, a series of deep gashes that lingers some 994 miles to the east of InSight's landing zone.

The results from this work are pending publication in a peer-reviewed journal, and scientists associated with the InSight team declined to comment until after the study's release. But the announcement of this active fault zone millions of miles away already has earthbound scientists abuzz. "All the expectations we have and all the models we have to try to explain how active Mars might be can now be benchmarked against this measurement," says Paul Byrne, a planetary geologist at North Carolina State University who is not part of the InSight team. "Mars has just become a bit more alive to us with these data."

Games have long served as a training ground for AI algorithms, and not without good reason. From a report: Games -- particularly video games -- provide challenging environments against which to benchmark autonomous systems. In 2013, a team of researchers introduced the Arcade Learning Environment, a collection of over 55 Atari 2600 games designed to test a broad range of AI techniques. More recently, San Francisco research firm OpenAI detailed Procgen Benchmark, a set of 16 virtual worlds that measure how quickly models learn generalizable skills. The next frontier might be Mega Man, if an international team of researchers have their way. In a newly published paper [PDF] on the preprint server Arxiv.org, they propose EvoMan, a game-playing competition based on the eight boss fights in Capcom's cult classic Mega Man 2. As they describe it, competitors' goal is to train an AI agent to defeat every enemy and evaluate their performances by common metrics.

Why Mega Man 2? The paper's coauthors assert that few other environments test an AI's ability to win against a single enemy, or how well an AI generalizes to win matches against waves of enemies or coevolves to create increasingly difficult enemies. To this end, in EvoMan, an AI-controlled Mega Man -- a robot equipped with a powerful arm cannon -- must beat eight so-called Robot Masters equipped with different weapons. Every time a Robot Master is defeated, the agent acquires its weapon, making it easier to defeat the bosses who remain. As proposed, EvoMan challenge entrants would train their agents on any four enemies and measure how well their learned strategy scales up to a whole set of enemies. The agents would be expected to learn how to identify and react to general patterns like avoiding being shot or shooting at the direction of the enemy, and to deplete an enemy's health from 100 energy points to 0 by the end of each match.

Japanese fashion retailer Uniqlo has managed to automate 90 percent of its flagship Tokyo warehouse, the Financial Times reports, adding that the facility includes a two-armed robot that can fold and box T-shirts, a job previously reserved for human staff. From a report: The industrial T-shirt folders come from Japanese robotics startup, Mujin. Uniqlo's parent company, Fast Retailing, partnered with Mujin to equip its flagship Tokyo warehouse with the robots, reducing its reliance on human labor in the process. FT says the innovation could enable the factory to completely automate production.

An anonymous reader shares a report: Whatever Foxconn is building in Wisconsin, it's not the $10 billion, 22 million-square-foot Generation 10.5 LCD factory that President Trump once promised would be the "eighth wonder of the world." At various points over the last two years, the Taiwanese tech manufacturer has said it would build a smaller LCD factory; that it wouldn't build a factory at all; that it would build an LCD factory; that the company could make any number of things, from screens for cars to server racks to robot coffee kiosks; and so on. Throughout these changes, one question has loomed: given that Foxconn is building something completely different than that Gen 10.5 LCD facility specified in its original contract with Wisconsin, is it still going to get the record-breaking $4.5 billion in taxpayer subsidies?

Documents obtained by The Verge show that Wisconsin officials have repeatedly -- and with growing urgency -- warned Foxconn that its current project has veered far from what was described in the original deal and that the contract must be amended if the company is to receive subsidies. Foxconn, however, has declined to amend the contract, and it indicated that it nevertheless intends to apply for tax credits. Foxconn has "refused by inaction" to amend the deal, says Wisconsin Department of Administration Secretary Joel Brennan. "They were continuously encouraged. It's a relatively recent development, where they have said, 'No, we don't want to do anything with the contract.' Our expectation has been, and continues to be, that they should want to come back and have discussions about this." Further reading: Wisconsin's $4.1 Billion Foxconn Boondoggle; One Year After Trump's Foxconn Groundbreaking, There is Almost Nothing To Show For It; and Foxconn is Confusing the Hell Out of Wisconsin.

A drone with a machine gun attached can hit targets with high precision, according to its makers. Turkey is set to become the first country to have the drone, when it gets a delivery this month. From a report: The 25-kilogram drone has eight rotating blades to get it in the air. Its machine gun carries 200 rounds of ammunition and can fire single shots or 15-round bursts. Many countries and groups already use small military drones that can drop grenades or fly into a target to detonate an explosive. The new drone, called Songar and made by Ankara-based electronics firm Asisguard, is the first drone to be equipped with a firearm and be ready for service. Turkey expects the drones to be delivered before the end of the year.

It is hard for a drone to shoot accurately, partly because of the difficulty of judging range and angle, and partly because the recoil from each shot significantly moves the drone, affecting the aim for the next round. Songar has two systems to overcome these challenges. One uses sensors, including cameras and a laser rangefinder, to calculate distance, angle and wind speed, and work out where to aim. The second is a set of robot arms that move the machine gun to compensate for the effects of recoil.

The AI world is booming in a range of metrics covering research, education, and technical achievements, according to AI Index report -- an annual rundown of machine learning data points now in its third year. From a news writeup, which outlines some of the more interesting and pertinent points: AI research is rocketing. Between 1998 and 2018, there's been a 300 percent increase in the publication of peer-reviewed papers on AI. Attendance at conferences has also surged; the biggest, NeurIPS, is expecting 13,500 attendees this year, up 800 percent from 2012.
AI education is equally popular. Enrollment in machine learning courses in universities and online continues to rise. Numbers are hard to summarize, but one good indicator is that AI is now the most popular specialization for computer science graduates in North America. Over 21 percent of CS PhDs choose to specialize in AI, which is more than double the second-most popular discipline: security / information assurance.
The US is still the global leader in AI by most metrics. Although China publishes more AI papers than any other nation, work produced in the US has a greater impact, with US authors cited 40 percent more than the global average. The US also puts the most money into private AI investment (a shade under $12 billion compared to China in second place globally with $6.8 billion) and files many more AI patents than any other country (with three times more than the number two nation, Japan).
AI algorithms are becoming faster and cheaper to train. Research means nothing unless it's accessible, so this data point is particularly welcome. The AI Index team noted that the time needed to train a machine vision algorithm on a popular dataset (ImageNet) fell from around three hours in October 2017 to just 88 seconds in July 2019. Costs also fell, from thousands of dollars to double-digit figures.
Self-driving cars received more private investment than any AI field. Just under 10 percent of global private investment went into autonomous vehicles, around $7.7 billion. That was followed by medical research and facial recognition (both attracting $4.7 billion), while the fastest-growing industrial AI fields were less flashy: robot process automation ($1 billion investment in 2018) and supply chain management (over $500 million).

An anonymous reader quotes Bloomberg: Robots have replaced thousands of routine jobs on Wall Street. Now, they're coming for higher-ups.

That's the contention of Marcos Lopez de Prado, a Cornell University professor and the former head of machine learning at AQR Capital Management LLC, who testified in Washington on Friday about the impact of artificial intelligence on capital markets and jobs. The use of algorithms in electronic markets has automated the jobs of tens of thousands of execution traders worldwide, and it's also displaced people who model prices and risk or build investment portfolios, he said.

"Financial machine learning creates a number of challenges for the 6.14 million people employed in the finance and insurance industry, many of whom will lose their jobs -- not necessarily because they are replaced by machines, but because they are not trained to work alongside algorithms," Lopez de Prado told the U.S. House Committee on Financial Services.

Japan's economy and industry ministry said Monday that it will start removing melted fuel from the Fukushima Daiichi nuclear power plant in 2021. The milestone step of debris removal is considered the most difficult part of cleaning up the crisis-hit facility. ABC News reports: Nearly nine years after [the Fukushima nuclear power plant was wrecked by a massive earthquake and tsunami], the decommissioning of the plant, where three reactors melted, remains largely an uncertainty. The revised road map, to be formally approved later this month, lacks details on how the complex should look at the end but maintains a 30- to 40-year target to finish.

By far the toughest challenge is to remove the 800 tons of nuclear fuel in the three reactors that melted, fell from the cores and hardened at the bottom of their primary containment vessels. In the past two years, plant operator Tokyo Electric Power Co. (TEPCO), has made progress in gathering details mainly from two of the three reactors. In February, a small telescopic robot sent inside Unit 2 showed that small pieces of debris can come off and be lifted out. The milestone step of debris removal is scheduled to begin at Unit 2 by the end of 2021. [...] TEPCO started removing the fuel rods from the Unit 3 pool in April 2019 and aims to get all 566 removed by March 2021. Removal of the rods from Units 1 and 2 is to begin in 2023. By 2031, TEPCO also plans to remove thousands at two other units that survived the tsunami to be stored in dry casks on the compound. More than 6,300 fuel rods were in six reactor cooling pools at the time of the accident, and only the Unit 4 pool has been emptied. "Japan has yet to develop a plan to dispose of the highly radioactive waste that will come out of the reactors," the report adds. "Under the road map, the government and TEPCO will compile a plan sometime after the first decade of debris removal ending in 2031."

"Experts say a 30- to 40-year completion target for the decommissioning is too optimistic. Some have raised doubts if removing all of the melted fuel is doable and suggest an approach like Chernobyl -- contain the reactors and wait until radioactivity naturally decreases."