Tag Archives: 2014
The brain has long inspired the design of computers and their software. Now Intel has become the latest tech company to decide that mimicking the brain’s hardware could be the next stage in the evolution of computing.
On Monday the company unveiled an experimental “neuromorphic” chip called Loihi. Neuromorphic chips are microprocessors whose architecture is configured to mimic the biological brain’s network of neurons and the connections between them called synapses.
While neural networks—the in vogue approach to artificial intelligence and machine learning—are also inspired by the brain and use layers of virtual neurons, they are still implemented on conventional silicon hardware such as CPUs and GPUs.
The main benefit of mimicking the architecture of the brain on a physical chip, say neuromorphic computing’s proponents, is energy efficiency—the human brain runs on roughly 20 watts. The “neurons” in neuromorphic chips carry out the role of both processor and memory which removes the need to shuttle data back and forth between separate units, which is how traditional chips work. Each neuron also only needs to be powered while it’s firing.
At present, most machine learning is done in data centers due to the massive energy and computing requirements. Creating chips that capture some of nature’s efficiency could allow AI to be run directly on devices like smartphones, cars, and robots.
This is exactly the kind of application Michael Mayberry, managing director of Intel’s research arm, touts in a blog post announcing Loihi. He talks about CCTV cameras that can run image recognition to identify missing persons or traffic lights that can track traffic flow to optimize timing and keep vehicles moving.
There’s still a long way to go before that happens though. According to Wired, so far Intel has only been working with prototypes, and the first full-size version of the chip won’t be built until November.
Once complete, it will feature 130,000 neurons and 130 million synaptic connections split between 128 computing cores. The device will be 1,000 times more energy-efficient than standard approaches, according to Mayberry, but more impressive are claims the chip will be capable of continuous learning.
Intel’s newly launched self-learning neuromorphic chip.
Normally deep learning works by training a neural network on giant datasets to create a model that can then be applied to new data. The Loihi chip will combine training and inference on the same chip, which will allow it to learn on the fly, constantly updating its models and adapting to changing circumstances without having to be deliberately re-trained.
A select group of universities and research institutions will be the first to get their hands on the new chip in the first half of 2018, but Mayberry said it could be years before it’s commercially available. Whether commercialization happens at all may largely depend on whether early adopters can get the hardware to solve any practically useful problems.
So far neuromorphic computing has struggled to gain traction outside the research community. IBM released a neuromorphic chip called TrueNorth in 2014, but the device has yet to showcase any commercially useful applications.
Lee Gomes summarizes the hurdles facing neuromorphic computing excellently in IEEE Spectrum. One is that deep learning can run on very simple, low-precision hardware that can be optimized to use very little power, which suggests complicated new architectures may struggle to find purchase.
It’s also not easy to transfer deep learning approaches developed on conventional chips over to neuromorphic hardware, and even Intel Labs chief scientist Narayan Srinivasa admitted to Forbes Loihi wouldn’t work well with some deep learning models.
Finally, there’s considerable competition in the quest to develop new computer architectures specialized for machine learning. GPU vendors Nvidia and AMD have pivoted to take advantage of this newfound market and companies like Google and Microsoft are developing their own in-house solutions.
Intel, for its part, isn’t putting all its eggs in one basket. Last year it bought two companies building chips for specialized machine learning—Movidius and Nervana—and this was followed up with the $15 billion purchase of self-driving car chip- and camera-maker Mobileye.
And while the jury is still out on neuromorphic computing, it makes sense for a company eager to position itself as the AI chipmaker of the future to have its fingers in as many pies as possible. There are a growing number of voices suggesting that despite its undoubted power, deep learning alone will not allow us to imbue machines with the kind of adaptable, general intelligence humans possess.
What new approaches will get us there are hard to predict, but it’s entirely possible they will only work on hardware that closely mimics the one device we already know is capable of supporting this kind of intelligence—the human brain.
Image Credit: Intel Continue reading
Today by far the most commonly used robotics software is ROS, which stands for Robot Operating System. This is an open source software, and the most number of developers and robotics users are involved with this program with an ever increasing rate. It contains set of libraries, algorithms, developer tools and drivers for developing robotics projects. The first release of ROS was in 2010, and as of end of 2016, ROS has reached its 10th official release, which is called “ROS Kinetic Kame”. There are translations to 11 languages other than English, which are: German, Spanish, French, Italian, Japanese, Turkish, Korean, Portuguese, Russian, Thai and Chinese. It currently has 2000+ software libraries, which keeps increasing every year.
Many robots use ROS now, including but not limited to hobby robots, drones, educational or advanced humanoid robots, domestic robots including cleaning robot vacuums, cooking robots or telepresence robots and more, robot arms, farming robots, industrial robots, even Robonaut of NASA in space or the four legged military robots in development. A list of robots which use ROS can be found here: http://wiki.ros.org/Robots.
We were checking the Alexa.Com ranking of ROS since few years, in order to track the increase in usage, and we believe it is time to share it now, as we have enough data. The numbers on the left are dates we looked and the numbers on the right indicate the ranking of Ros.Org website from top, among all websites in the world:
May 2011: 189,000 th in the world, from top, among all other websites
April 2012: 187,900 th
January 2014: 107,821
May 2014: 112,236
September 2014: 83,875 (7219 in Canada, the country where it is most accessed)
January 2015: 83,556 (4,258 in Canada)
February 2015 : 75,680 (33185 in USA)
April 2015: 59,200 (31,334 in USA)
August 2015: 65,754 (50,132 in USA)
September 2016: 30,201 (China 5073)
This chart shows the increasing rank of ros.org among other websites in the world, which is a good indicator of its growth. The numbers on the left represent the site’s ranking from the top, among all other sites in the world. Chart Copyright: Robokingdom LLC.
As can be seen here, in May 2011, when we first checked this ranking, ROS.org was at 189,000 th place in the world from the top among all other websites in terms of unique visitors that visit the site, and it almost continuously increased its ranking. As of September 2016, it is now the 30,201st most reached website in the world, with mostly being accessed in China (5073 from top in China). Let’s not forget that even if it’s position remained the same, let alone going up, it would still mean the traffic of the site was going up, as every year there are more websites in the world which means the same ranking means better place and more traffic. The ranking of 30,201 means ROS.org is a very high traffic website in the world right now, being accessed probably by at least hundreds of thousands of people every day, with no indication of slowing down its rise yet.
The most important result of all of this, is that the use of robots is increasing, both in terms of number and type (when you look at the type of robots that use ros, as it also increases in variety all the time).
From Alexa, we were also able to see, from publicly available information, that the percentage of reach among countries for ROS.org is as follows:
South Korea 3.5%
This also shows us that in China, a lot of things are going on for robotics development right now, as it gets most of its traffic from there with 47.5%. USA then follows with 11.5% and Japan is third with 8.7%.
With ROS, any type of sensors can be controlled, including 1d/2d range sensors, 3d range finders and cameras, audio/speech recognition sensors, cameras, environmental sensors, force/torque/touch sensors, motion capture, pose estimation, power supply, RFID, and sensor interfaces.
In ros.org site, in addition to all packages, there are also extensive tutorials and a discussion board that one can ask questions and share knowledge.
ROS also has an industrial section, the version of software modified for industrial applications. It is called ROS industrial, and can be reached at: http://rosindustrial.org/. Although we see domestic robots with new abilities or advanced research projects that aim to develop capabilities of robotics every year, according to the results of a study that is shown on http://rosindustrial.org/the-challenge/ website, the abilities of industrial robots are not progressing and the abilities are restricted to welding, material handling, dispensing, coating (although we know that they do additional tasks such as packaging, inspection, labeling etc…). ROS Industrial aims to solve this challenge by providing a common skeleton to all developers, with its extensive and stronger software architecture, than other individual robotics programs.
The post Robot Operating System (ROS) continues its growth appeared first on Roboticmagazine. Continue reading