System Bits: July 15

Automating bridge inspections with robotics
The University of Waterloo has come up with robotics that could be used in automated inspection of bridges, making sure such critical infrastructure is safe and sound. The technology promises to make bridge inspection cheaper and easier.

The system collects data for defect detection and analysis through a combination of autonomous robots, cameras, and LiDAR technology, which is incorporated in self-driving cars and in other applications, such as mapping.

“We can do more than humans now do – and do it much better in every way,” said Sriram Narasimhan, an engineering professor at Waterloo. “It is very inexpensive because you don’t need as many inspectors relying on specialized equipment, such as lifts, and you get much higher quality information.”

Narasimhan, a Canada Research Chair in Smart Infrastructure, said current practices create an inspection system that is subjective, less repeatable and often imprecise because it is based, at least in part, on educated guesswork.

The automated system, by contrast, eliminates the subjectivity of human inspectors, and is both repeatable and reliable, with the ability to precisely measure the size of defects and reveal invisible, sub-surface problems with infrared cameras.

It is designed so that the results from one inspection can be overlaid on previous inspection results on a detailed map displaying dozens of key vulnerable areas of the subject bridge.

Stephen Phillips, a PhD student at the University of Waterloo, works with a robotic ground vehicle used in research on automating bridge inspections. Photo credit: Christopher Bogdon, Clearpath Robotics

“The benefit is that we can track and quantify defects as they evolve over time,” said Narasimhan, director of the Structural Dynamics Identification and Control Laboratory. “That is not practically possible with humans alone, but it is with the assistance of robots.”

A wheeled ground vehicle used in the research was programmed with an inspection plan detailing instructions on its location and areas of the bridge to focus on.

The same software could also be used for inspections with water-going vehicles and airborne drones, or to inspect other infrastructure such as nuclear power plants and buildings.

Researchers are now developing water-borne inspection platforms and artificial intelligence algorithms to automatically detect and identify particular kinds of defects.

Further down the road, they hope to automate analysis of the data collected by robots to much more accurately assess the structural integrity of infrastructure and forecast the need for repairs or replacement.

“We’re combining our infrastructure expertise with the latest robotics technology to greatly improve what is now a very manual inspection process,” Narasimhan said.

Dealing with presbyopia through autofocals
Presbyopia is a vision defect that afflicts many people at the age of 45 and older. Reading glasses are sufficient for some people, while others usually resort to progressive lenses as the main alternative to eye surgery.

At Stanford University, researchers have created new lenses that account for presbyopia better than progressive lenses do.

“More than a billion people have presbyopia and we’ve created a pair of autofocal lenses that might one day correct their vision far more effectively than traditional glasses,” said Stanford electrical engineer Gordon Wetzstein. For now, the prototype looks like virtual reality goggles, but the team hopes to streamline later versions.

Wetzstein’s prototype glasses – dubbed autofocals – are intended to solve the main problem with today’s progressive lenses: These traditional glasses require the wearer to align their head to focus properly. Imagine driving a car and looking in a side mirror to change lanes. With progressive lenses, there’s little or no peripheral focus. The driver must switch from looking at the road ahead through the top of the glasses, then turn almost 90 degrees to see the nearby mirror through the lower part of the lens.

This visual shift can also make it difficult to navigate the world. “People wearing progressive lenses have a higher risk of falling and injuring themselves,” said graduate student Robert Konrad, a co-author on a paper describing the autofocal glasses published June 28 in the Science Advances journal.

The Stanford prototype works much like the lens of the eye, with fluid-filled lenses that bulge and thin as the field of vision changes. It also includes eye-tracking sensors that triangulate where a person is looking and determine the precise distance to the object of interest. The team did not invent these lenses or eye-trackers, but they did develop the software system that harnesses this eye-tracking data to keep the fluid-filled lenses in constant and perfect focus.

Nitish Padmanaban, a graduate student and first author on the paper, said other teams had previously tried to apply autofocus lenses to presbyopia. But without guidance from the eye-tracking hardware and system software, those earlier efforts were no better than wearing traditional progressive lenses.

To validate its approach, the Stanford team tested the prototype on 56 people with presbyopia. Test subjects said the autofocus lenses performed better and faster at reading and other tasks. Wearers also tended to prefer the autofocal glasses to the experience of progressive lenses – bulk and weight aside.

If the approach sounds a bit like virtual reality, that isn’t far off. Wetzstein’s lab is at the forefront of vision systems for virtual and augmented reality. It was in the course of such work that the researchers became aware of the new autofocus lenses and eye-trackers and had the insight to combine these elements to create a potentially transformative product.

The next step will be to downsize the technology. Wetzstein thinks it may take a few years to develop autofocal glasses that are lightweight, energy efficient and stylish. But he is convinced that autofocals are the future of vision correction.

“This technology could affect billions of people’s lives in a meaningful way that most techno-gadgets never will,” he said.

This research was funded in part by Intel, Nvidia, an Okawa Research Grant, a Sloan Fellowship, and the National Science Foundation.

Magnetic monopoles sound off
Here’s a scientific advance, not found in jellyfish, but in a SQUID.

Magnetic monopoles are fundamentally important but highly elusive elementary particles exhibiting quantized magnetic charge. The prospect for studying them has brightened in recent years with the theoretical realization that, in certain classes of magnetic insulators, the thermally excited states exhibit all the characteristics of magnetic monopoles.

Now, a collaboration led by Professor J.C. Séamus Davis and Professor Stephen J. Blundell of the University of Oxford’s Department of Physics has developed a new approach to detecting and studying these “emergent” magnetic monopoles – including the discovery that, when amplified, the noise they make is audible to humans. The findings are published in the journal Nature.

In 2018, Professor Blundell and his colleagues Dr. Franziska Kirschner and Dr. Felix Flicker predicted that the random motion of magnetic monopoles inside these compounds would generate a very specific kind of magnetization noise.

This means that a crystal of one of these magnetic insulators should spontaneously generate wildly and randomly fluctuating magnetic fields both internally and externally, as the monopoles move around. The catch was that these fields vary rapidly and randomly at every point, so that the net fluctuating field through a sample was predicted to be near one-billionth of the Earth’s field.

In response, Professor Davis and colleague Dr. Ritika Dusad built an exquisitely sensitive magnetic-field-noise spectrometer based on a superconducting quantum interference device – a SQUID.

Professor Davis said, “Virtually all the predicted features of the magnetic noise coming from a dense fluid of magnetic monopoles were then discovered emerging from crystals of Dy2Ti2O7. Extraordinarily, because this magnetic monopole noise occurs in the frequency range below 20kHz, when amplified by the SQUID it is actually audible to humans.”

Professor Blundell added, “What makes magnetic monopoles fascinating is that they ‘emerge’ from a dense lattice of magnetic monopoles, and this makes their motion highly constrained – very different from a typical gas of charged particles. This observation led us on a search for the signature of this constrained motion in the magnetic noise spectrum. These exciting results open up the possibility of using magnetic noise to study many other exotic magnetic systems containing different species of emergent particles.”