Vic Bishop, Staff WriterResearch into the structure and function of the human brain continues to accelerate. Collaborations, such as the Human Brain Project in Europe and the BRAIN initiative in the United States, are exploring making great advances in understanding the brain’s circuitry and computing principles.The supposed goals of these research initiatives are to understand the cause of and to improve treatment of brain disorders, to create neu [...]
After years of silence on all but the most prosaic aspects of the secretive X-37B space plane program, the Defense Department has revealed that the mysterious, truck-sized craft's next mission will host an experimental new thrust system that could greatly improve the shelf life of satellites.
The X-37B program has sent its shuttle-like Orbital Test Vehicle craft into space three times for a total time in orbit of almost four years. What the spacecraft has been doing up there is anybody's guess — its creators have declined to comment except to say that everything is working properly. But a news release this week from the Air Force says in no uncertain terms that the next flight of the X-37B, set to begin next month, will be the platform for testing a Hall thruster.
Hall thrusters combine electricity and a noble gas like xenon to produce a miniscule amount of direct force — weak in comparison with thrusters that use ordinary solid fuel, but at a far lesser cost of fuel. Trading power for fuel efficiency would allow satellites and probes to make course adjustments for much longer, extending their lives and versatility. Spaceflight Now has more details on how the system works.
Of course, this sheds no light on what the last three X-37B missions were — but in light of this new information it seems more likely that it's a test bed for high-tech space experiments, and not an orbital bomber or elite spy satellite. But you never know.
pcmag.com IBM's Watson supercomputer may be saving lives and educating children, but Google's new AI program can master video games without human guidance.
The artificial intelligence system from London-based DeepMind, which Google acquired last year for a reported $400 million, represents a major step toward a future of smart machines.
Computers running the deep Q-network (DQN) algorithm were exposed to 49 retro games on the Atari 2600 and told to play them, without any direction from researchers. Using the same network architecture and tuning parameters, the machines were given only raw screen pixels, available actions, and game score as input.
For each level passed or high score earned, the computer was automatically rewarded with a digital treat.
"Strikingly, DQN was able to work straight 'out of the box' across all these games," DeepMind's Dharshan Kumaran and Demis Hassabis wrote in a blog post. The executives cited classic titles like Breakout, River Raid, Boxing, and Enduro.
The AI crushed even the most expert humans at 29 games, sometimes composing what the creators called "surprisingly far-sighted strategies" that allowed maximum scoring possibilities. It also outperformed previous machine-learning methods in 43 of 49 instances.
VIEW ALL PHOTOS IN GALLERY
Google DeepMind's findings were presented in a paper published in this week's Nature journal, which describes the key DQN features that allow it to learn.
"This work offers the first demonstration of a general purpose learning agent that can be trained end-to-end to handle a wide variety of challenging tasks," the researchers said. "This kind of technology should help us build more useful products."
Imagine asking the Google app to complete a complex task—like plan a backpacking trip through Europe, for example.
Google's DeepMind also hopes its technology will give researchers new ways to make sense of large-scale data, opening the door to discoveries in fields like climate science, physics, medicine, and genomics.
"And it may even help scientists better understand the process by which humans learn," Kumaran and Hassabis said, citing physicist Richard Feynman, who famously said, "What I cannot create, I do not understand."
For more, see How DeepMind Can Bring Google Artificial Intelligence to Life in the slideshow above.
IT'S been described as science fiction made real - but now, just as the final selection process gets under way for the folk with the right stuff to make a manned mission to Mars, scientists have dashed the dreams of planet Earth by warning the journey will probably never happen and will end in disaster if it does.
Privately run space exploration programme Mars One wants to send four people to the red planet for the rest of their (probably not very long) lives and film it for reality TV in order to help finance the endeavour.
Thousands have set their sights on becoming the first settlers to land on the planet - and have now been whittled down to a short list of 100, including a Scottish PhD student - but with questionable technology, a lack of funding and an unrealistic timeframe, experts claim it is a "suicide mission".
Mars One believes it can achieve a manned mission in 2024 - sooner than NASA, the European Space Agency, the Russians or Chinese, and on a fraction of their budgets.
If the project does go ahead, the crew would have to make it through nine months of interplanetary travel without being killed by mishap, radiation - or each other.
And even then, a recent study suggested they will only last 68 days on Mars before dying - due to lack of food and water.
However, Anu Ojha OBE, director of the UK National Space Academy Programme, has warned the applicants not to get their hopes up as the mission is unlikely to ever leave the ground.
Ojha said: "Obviously this is something that has captured the public's imagination, and Mars One obviously has a great PR team, but space engineering obeys the laws of physics not PR." Mars One is the brainchild of Dutch entrepreneur Bas Lansdorp who was inspired by the images of Mars sent back by the Sojourner rover in 1997, when he was a student.
Lansdorp, who will not make the journey himself, has an impressive team working on the project including former NASA employees Dr Norbert Kraft, who specialises in the physiological and psychological effects of space travel and space architect Kristian von Bengtson.
Physicist Arno Wielders, who previously worked for Dutch Space, is also on board, as well as a number of other advisers from around the world with backgrounds in space engineering, science and technology, marketing, design and television production.
The ultimate aim is to see a large, self-sustaining colony on Mars, but Ojha, who is also a director at the National Space Centre in Leicester, said there are three major stumbling blocks for the mission: technology, funding and human psychology.
"In terms of technology, it's pushing the absolute boundaries and there seems to be a lot of technological naivety on the part of the people running it", he said.
"There are some elements that seem reasonable, but overall it's concerning, and the timescales are also questionable."
While Mars One is planning the one way mission for 2024, NASA, with its long established expertise and technology, is looking to be able to send humans to Mars and bring them back again by the mid 2030s.
This is estimated to cost up to as much as £100 billion (£64.9bn) for the space agency, while Mars One believes it can do it for an optimistic $6 billion (£3.9bn) - and there are even questions over whether or not they will be able to achieve that much funding. The private enterprise is hoping to raise money through a TV deal and additional funding from the exposure that will bring the project.
Last year it said it had teamed up with programme makers Endemol, but the Big Brother creators recently pulled out of the deal claiming they were "unable to reach agreement on the details of the contract".
Mars One did not respond to questioning by the Sunday Herald over its funding, but its website showed that as at January this year, it had raised just $759,816 from donations, merchandising, and a crowdfunding campaign.
It is unclear what other funding the project has.
Ojha said: "The business model has so many holes in it, it's shaky to say the least. And when you ask them how much money they have raised, they say it's still ongoing. The time scales and the business model - they're completely unrealistic."
Mars One plans to send several unmanned rockets to Mars ahead of the 2024 mission, with the first of these scheduled to take place in 2018.
These will include missions with robots to find a suitable location for a base and assemble it ahead of the humans' arrival. The project claims it will use only existing technology for the mission, buying in materials from proven suppliers including Lockheed Martin or SpaceX.
The equipment involved includes several simulation outposts for training, a rocket launcher, a transit vehicle to take the crew to Mars, a Mars landing capsule, two rovers, a Mars suit and a communications system.
However, experts have warned that much of this equipment has not been fully tested.
Physicist professor Todd Huffman is a big supporter of attempting a manned mission to Mars, but he also has serious concerns about Mars One, claiming it is "scientifically irresponsible".
He said: "The plan stretches the technology in many places. "The launch vehicle they want to use has not actually ever launched yet, let alone make a trip to Mars.
"The living spaces have not been made nor has it been tested whether they can be robotically assembled and by what kind of robot.
"A suitable site would also need to be found for the living spaces and the details of how water extraction will take place have not been understood.
"If you assign a 90 per cent chance to success to each of those things, all of which are necessary for human survival, you end up with about a 50 per cent chance of failure, ending in the death of the colonists - and that would likely not make good television." He added: "Unless we [wait for] quite a lot of technology and exploration to happen first, it is basically worse than a one-way ticket for the colonists - it is almost surely a suicide mission if carried out within this next decade."
Although most scientists believe the mission will not go ahead, some have also warned of the psychological impact on the people selected for the mission if it does.
Ojha said: "The thing that's really captured the public's imagination is this idea of it being a one way trip, but this brings another set of problems in terms of human psychology.
"The longest period a human has spent in space is 438 days - they're talking about sending people on a one way trip. "Lots of the people I've seen interviewed, they're really excited about taking part, but have they really thought about what they're doing and what the implications are?
"I would tell them to go to Antarctica for six months in the middle of winter and that's about 1 per cent of what they'll be experiencing on Mars.
"Human psychology is far more fragile than we think."
However, while many scientists warn of the dangers and do not believe the mission will proceed, they have praised Mars One for sparking the public's interest in planetary science.
Dr John Bridges, of the Space Research Centre in Leicester, said: "It's a very interesting and innovative project, but the time scales are very challenging.
"I believe they're planning for 2024 and it's 2015 now. So for something as major as this, it's a very challenging timescale "But it's fantastic that people are thinking about this, that industry is getting involved and raising awareness of planetary science."
Ojha added: "Mars One has been great in a way because it's once again drawn people's imagination to the idea of space engineering and exploration.
"But the reality is that there are serious concerns about the project's space engineering, funding and medical implications."
Lansdorp has previously said that most people are "surprised to hear that the manned missions will be happening in ten years time, with a budget ten times less than Nasa".
He added: "But I think that if you really spend time studying Mars One, you cannot believe there is not a good chance we will make it. "At the same time, it's a hugely ambitious plan, there's many things that can go wrong with such a big plan.
"But I believe we have a good plan and we can overcome the challenges."
However, he has also conceded that the current plans are an "optimum schedule", adding: "If one rocket doesn't launch, or a lander doesn't work on Mars before a human goes, any major malfunctions will result in a two year delay."
Mars One declined the Sunday Herald's request to interview someone from the project and failed to answer any of our questions.