Top Tech Day 10: Quantum Computers & Humanoid Robotics
Unfortunately we missed posting a technology yesterday due to a technical failure here at the office (primarily, our power was out due to the snow). So today we’re covering TWO technologies, and we’re picking some exciting ones at that. The technologies we’re updating you on today are quantum computing, and humanoid robotics.
Quantum computing is a technology that has dazzled the minds of futurists for years-and for good reason. Being able to address certain computational problems that are currently intractable or effectively intractable is alluring, and different types of quantum computers do just that. You can expect a more thorough writeup on them from us at some point in the future regardless, but for the point of this article, we’ll note that there’s a difference between the quantum computing currently most discussed in the news (D-Wave’s adiabatic quantum annealer) and what we’re talking about today. The qubits discussed today are not yet in a quantum computer, but would be more appropriate for use in one of the other types of quantum computer. While the full scale of the applications of such a system would take significantly more space and time than we have available here today, quantum computers are useful for cryptography, mathematics, physics and chemistry simulations, and any set of problems that has the properties:
- The only way to solve the problems is to guess answers and check them
- The number of answers is proportional to number of inputs
- Every possible answer takes the same time to check
- There is no benefit to a particular order of checking answers (ie random is as good as sorted).
There are two major advances we wanted to discuss today-one on the lifetime of entangled qubits, and one on the number of entangled qubits.
Australian scientists at the University of Sydney, Australian National University, and Japanese scientists at the University of Tokyo have had a breakthrough in the number of qubits created. The previous record for entangled qubits, set in 2011, was 14. The team of scientists led by Dr Nicolas Menicucci (a world renowned quantum computation expert) managed to create a ‘continuous variable clustered state’ containing more than 10,000 modes. In addition, the paper also laid out a method of using this system for quantum computation using sequential applications of quantum teleportation. The researchers said that further breakthroughs in precise control of the devices are required to take advantage of the increased scale. Further information can be found in the paper itself, here.
The other major advance was the increased longevity of a quantum bit. A team at Simon Fraser University managed to create a qubit stable at room temperature for 39 minutes, as well as remaining stable after repeated heating and cooling processes. Information was only able to be read at under 10 kelvin, but this is a significant advance over previous efforts. The next step for the researchers is to allow multiple qubits, as well as reading at room temperature. More information can be found here.
Our second topic today is humanoid robotics. Humanoid robotics is one area of robotics that receives a huge amount of attention-not only does it fulfill some of the greatest dreams of futurists and sci-fi fans alike, but it also invokes some of their deepest fears. In the ideal case for developers, humanoid robots would be able to do everything a human could do (at least in their area of specialization), and every year we get closer to achieving that dream…though it’s still a ways off.
There are three articles of note this year that we feel capture how humanoid robotics is advancing. DARPA’s Robotics Challenge (DRC) will be taking place on the 20th and 21st of this month, where robots designed by a number of schools and institutes will be competing on emergency worker tasks. Tasks include driving a vehicle, navigating complicated terrain, climbing a ladder, moving debris, opening a door, cutting through a wall, shutting off a valve, and moving a hose into place. The ‘A Track’ teams received funding from DARPA and developed their own robots. ‘B Track’ teams received funding from DARPA and competed in a software only competition with unfunded ‘C Track’ teams for the rights to a limited number of ATLAS robots (developed by Boston Dynamics). ‘D Track’ teams produced their own robots and software, unfunded. A total of 17 teams will be competing at the DRC later this month.
INRIA Flowers, a robotic team at the French research institute, have used 3d printing to build a miniaturized humanoid robot for only $11,000 including servos and electronics. The 33 inch, 7.7 pound robot is powered by a Raspberry Pi and equipped with force sensing resistors, two cameras, a stereo microphone, an inertial measurement unit, and an LCD for a face. Custom designed from the ground up to be bio-inspired, it has a number of uncommon design choices (an articulated spine, additional leg springs, heel to toe gait) to enable the researchers to study bipedal walking. More can be found at the Flowers lab and Poppy Project homepage.
The final item for today’s blog post is news from just last week that some may not be familiar with-Google is getting into the robotics market. They’ve made 7 core acquisitions in the last year: Industrial Perception (computer vision), Autofuss (robot enabled cinematography), Bot & Dolly (robot enabled cinematography), Schaft (humanoid robotics), Meka (possibly humanoid robotics), Redwood Robotics (possibly robotic arms), and Holomni (omnidirectional robotic wheels). Powered by these technologies, Andy Rubin (former head of Android) has been tasked with leading a division to design and build a fleet of Googlebots. What for? Everything in the Google horizontal-from manufacturing, to packing, to home delivery. More information can be found at the New York Times and Singularity Hub.