Robot Vs. Superbug

November 30, 2019 by Bryan Cockfield 6 Comments

Working in a university or research laboratory on interesting, complicated problems in the sciences has a romanticized, glorified position in our culture. While the end results are certainly worth celebrating, often the process of new scientific discovery is underwhelming, if not outright tedious. That’s especially true in biology and chemistry, where scaling up sample sizes isn’t easy without a lot of human labor. A research group from Reading University was able to modify a 3D printer to take some of that labor out of the equation, though.

This 3D printer was used essentially as a base, with the printing head removed and replaced with a Raspberry Pi camera. The printer X/Y axes move the camera around to all of the different sample stored in the print bed, which allows the computer attached to the printer to do most of the work that a normal human would have had to do. This allows them to scale up massively and cheaply, presumably with less tedious inputs from a large number of graduate students.

While the group hopes that this method will have wide applicability for any research group handling large samples, their specific area of interest involves researching “superbugs” or microbes which have developed antibiotic resistance. Their recently-published paper states that any field which involves bacterial motility, colony growth, microtitre plates or microfluidic devices could benefit from this 3D printer modification.

Your WiFi Signals Are Revealing Your Location

November 28, 2019 by Sharon Lin 43 Comments

The home may be the hearth, but it’s not going to be a place of safety for too long.

With the abundance of connected devices making their ways into our homes, increasing levels of data may allow for more accurate methods for remote surveillance. By measuring the strength of ambient signals emitted from devices, a site can be remotely monitored for movement. That is to say, WiFi signals may soon pose a physical security vulnerability.

In a study from the University of Chicago and the University of California, Santa Barbara, researchers built on earlier studies where they could use similar techniques to “see through walls” to demonstrate a proof-of-concept for passive listening. Attackers don’t need to transmit signals or break encryptions to gain access to a victim’s location – they just need to listen to the ambient signals coming from connected devices, making it more difficult to track bad actors down.

Typically, connected devices communicate to an access point such as a router rather than directly with the Internet. A person walking near a device can subtly change the signal propagated to the access point, which is picked up by a receiver sniffing the signal. Most building materials do not block WiFi signals from propagating, allowing receivers to be placed inconspicuously in different rooms from the access point.

WiFi sniffers are relatively inexpensive, with models running for less than $20. They’re also small enough to hide in unsuspecting locations – inside backpacks, inside a box – and emit no signal that could be detected by a target. The researchers proposed some methods for safeguarding against the vulnerability: insulating buildings against WiFi leakage (while ensuring that desirable signals, i.e. signals from cell tower are still able to enter) or having access points emit a “cover signal” that mixes signals from connected devices to make it harder to sniff for motion.

While we may not be seeing buildings surrounded by Faraday cages anytime soon, there’s only going to be more attack surfaces to worry about as our devices continue to become connected.

[Thanks to Qes for the tip!]

A Single-Digit-Micrometer Thickness Wood Speaker

November 27, 2019 by Sharon Lin 22 Comments

Researchers have created an audio speaker using ultra-thin wood film. The new material demonstrates high tensile strength and increased Young’s modulus, as well as acoustic properties contributing to higher resonance frequency and greater displacement amplitude compared to a commercial polypropylene diaphragm in an audio speaker.

Typically, acoustic membranes have to remain very thin (on the micron scale) and robust in order to allow for a highly sensitive frequency response and vibrational amplitude. Materials made from plastic, metal, ceramic, and carbon have been used by engineers and physicists in an attempt to enhance the quality of sound. While plastic thin films are most commonly manufactured, they have a pretty bad impact on the environment. Meanwhile, metal, ceramic, and carbon-based materials are more expensive and less attractive to manufacturers as a result.

Cellulose-based materials have been making an entrance in acoustics research with their environmentally friendly nature and natural wooden structure. Materials like bagasse, wood fibers, chitin, cotton, bacterial cellulose, and lignocellulose are all contenders for effective alternatives to parts currently produced from plastics.

The process for building the ultra-thin film involved removing lignin and hemicellulose from balsa wood, resulting in a highly porous material. The result is hot pressed for a thickness reduction of 97%. The cellulose nano-fibers remain oriented but more densely packed compared to natural wood. In addition, the fibers required higher energy to be pulled apart while remaining flexible and foldable.

At one point in time, plastics seemed to be the hottest new material, but perhaps wood is making a comeback?

[Thanks Qes for the tip!]

Robotic Skin Sees When (and How) You’re Touching It

November 25, 2019 by Sharon Lin 2 Comments

Cameras are getting less and less conspicuous. Now they’re hiding under the skin of robots.

A team of researchers from ETH Zurich in Switzerland have recently created a multi-camera optical tactile sensor that is able to monitor the space around it based on contact force distribution. The sensor uses a stack up involving a camera, LEDs, and three layers of silicone to optically detect any disturbance of the skin.

The scheme is modular and in this example uses four cameras but can be scaled up from there. During manufacture, the camera and LED circuit boards are placed and a layer of firm silicone is poured to about 5 mm in thickness. Next a 2 mm layer doped with spherical particles is poured before the final 1.5 mm layer of black silicone is poured. The cameras track the particles as they move and use the information to infer the deformation of the material and the force applied to it. The sensor is also able to reconstruct the forces causing the deformation and create a contact force distribution. The demo uses fairly inexpensive cameras — Raspberry Pi cameras monitored by an NVIDIA Jetson Nano Developer Kit — that in total provide about 65,000 pixels of resolution.

Apart from just providing more information about the forces applied to a surface, the sensor also has a larger contact surface and is thinner than other camera-based systems since it doesn’t require the use of reflective components. It regularly recalibrates itself based on a convolutional neural network pre-trained with data from three cameras and updated with data from all four cameras. Possible future applications include soft robotics, improving touch-based sensing with the aid of computer vision algorithms.

While self-aware robotic skins may not be on the market quite so soon, this certainly opens the possibility for robots that can detect when too much force is being applied to their structures — the machine equivalent sensation to pain.

Continue reading “Robotic Skin Sees When (and How) You’re Touching It” →

Incredibly Tiny RF Antennas For Practical Nanotech Radios

November 25, 2019 by Sharon Lin 18 Comments

Researchers may have created the smallest-ever radio-frequency antennas, a development that should be of interest to any nanotechnology enthusiasts. A group of scientists from Korea published a paper in ACS Nano that details the fabrication of a two-dimensional radio-frequency antenna for wearable applications. Most antennas made from metallic materials like aluminum, cooper, or steel which are too thick to use for nanotechnology applications, even in the wearables space. The newly created antenna instead uses metallic niobium diselenide (NbSe2) to create a monopole patch RF antenna. Even with its sub-micrometer thickness (less than 1/100 the width of a strand of human hair), it functions effectively.

The metallic niobium atoms are sandwiched between two layers of selenium atoms to create the incredibly thin 2D composition. This was accomplished by spray-coating layers of the NbSe2 nanosheets onto a plastic substrate. A 10 mm x 10 mm patch of the material was able to perform with a 70.6% radiation efficiency, propagating RF signals in all directions. Changing the length of the antenna allowed its frequency to be tuned from 2.01-2.80 GHz, which includes the range required for Bluetooth and WiFi connectivity.

Within the ever-shrinking realm of sensors for wearable technologies, there is sure to be a place for tiny antennas as well.

[Thanks Qes for the tip!]

Qantas’ Research Flight Travels 115% Of Range With Undercrowded Cabin

November 6, 2019 by Lewin Day 34 Comments

Long-haul flights can be a real pain when you’re trying to get around the world. Typically, they’re achieved by including a stop along the way, with the layover forcing passengers to deplane and kill time before continuing the flight. As planes have improved over the years, airlines have begun to introduce more direct flights where possible, negating this frustration.

Australian flag carrier Qantas are at the forefront of this push, recently attempting a direct flight from New York to Sydney. This required careful planning and preparation, and the research flight is intended to be a trial run ahead of future commercial operations. How did they keep the plane — and the passengers — in the air for this extremely long haul? The short answer is that they cheated with no cargo and by pampering their 85% empty passenger cabin. Yet they plan to leverage what they learn to begin operating 10,000+ mile non-stop passenger flights — besting the current record by 10% — as soon as four years from now.
Continue reading “Qantas’ Research Flight Travels 115% Of Range With Undercrowded Cabin” →

An Algorithm For De-Biasing AI Systems

October 21, 2019 by Sharon Lin 52 Comments

A fundamental truth about AI systems is that training the system with biased data creates biased results. This can be especially dangerous when the systems are being used to predict crime or select sentences for criminals, since they can hinge on unrelated traits such as race or gender to make determinations.

A group of researchers from the Massachusetts Institute of Technology (MIT) CSAIL is working on a solution to “de-bias” data by resampling it to be more balanced. The paper published by PhD students [Alexander Amini] and [Ava Soleimany] describes an algorithm that can learn a specific task – such as facial recognition – as well as the structure of the training data, which allows it to identify and minimize any hidden biases.

Testing showed that the algorithm minimized “categorical bias” by over 60% compared against other widely cited facial detection models, all while maintaining the same precision of detection. This figure was maintained when the team evaluated a facial-image dataset from the Algorithmic Justice League, a spin-off group from the MIT Media Lab.

The team says that their algorithm would be particularly relevant for large datasets that can’t easily be vetted by a human, and can potentially rectify algorithms used in security, law enforcement, and other domains beyond facial detection.