Beyond biomimicry: Scientists find better-than-nature run style for six-legged robots

Researchers have found a running style for six-legged robots that significantly improves on the traditional nature-inspired method of movement.

The research, conducted by scientists at the École Polytechnique Fédérale de Lausanne (EPFL) and the University of Lausanne (UNIL) in Switzerland, found that as long as the robots are not equipped with insect-like adhesive pads, it is faster for them to move with only two legs on the ground at any given time.

Robotics has in the past few years made heavy use of biomimicry – the practice of mimicking natural systems – resulting in six-legged robots being designed to move like insects. In nature, insects use what is known as a tripod gait, where they have three legs on the ground at a time, so it had been assumed that this was the most efficient way for similarly legged robots to move.

However, by undertaking a series of computer simulations, tests on robots and experiments on Drosophila melanogaster – better known as the common fruit fly – the scientists found that the two-legged approach, which they have dubbed the bipod gait, results in faster and more efficient movement.

The core goal of the research, which is published today in the journal Nature Communications, was to confirm whether the long-held assumption that a tripod gait was best was indeed correct.

“We wanted to determine why insects use a tripod gait and identify whether it is, indeed, the fastest way for six-legged animals and robots to walk,” said Pavan Ramdya, study co-lead and corresponding author.

Initially, this involved the use of a simulated insect model based on the common fruit fly and an algorithm designed to mimic different evolutionary stages. This algorithm simulated different potential gaits to create a shortlist of those that it deemed to be the fastest.

This, however, shed light on why insects have a tripod gait – and why it may not be the best option for robots. The simulations showed that the traditional tripod gait works in combination with the adhesive pad found on the ends of insects’ legs to make climbing over vertical surfaces such as rocks easier and quicker.

Robots, however, are typically designed to walk along flat surfaces, and so the benefits of such a gait are lost.

“Our findings support the idea that insects use a tripod gait to most effectively walk on surfaces in three dimensions, and because their legs have adhesive properties. This confirms a long-standing biological hypothesis,” said Ramdya. “Ground robots should therefore break free from only using the tripod gait”.

Study co-lead authors Robin Thandiackal (left) and Pavan Ramdya with the six-legged robot used in the research. Images courtesy of EPFL/Alain Herzog

To for always corroborate the simulation’s findings, the researchers built a six-legged robot that could move either with a bipod or tripod gait, and which quickly confirmed the research by being faster when moving with just two legs on the ground at once.

However, they went further by confirming that the adhesive pads were in fact playing a role in the insect’s tripod movement.

They did this by equipping the fruit flies with tiny polymer boots that would cover the adhesive pads, and so remove their role in the way the insects moved. The flies’ responses confirms their theory: they began moving with a bipod-like gate rather than their conventional tripod-style movement.

“This result shows that, unlike most robots, animals can adapt to find new ways of walking under new circumstances,” said study co-lead author Robin Thandiackal.

As bizarre as the research sounds, it provides valuable new insights both for roboticists and biologists, and could lead to a new standard in the way that six legged robots are designed to move.

“There is a natural dialogue between robotics and biology: Many robot designers are inspired by nature and biologists can use robots to better understand the behavior of animal species,” added Thandiackal. “We believe that our work represents an important contribution to the study of animal and robotic locomotion.”

School will use facial analysis to identify students who are dozing off

In September the ESG business school in Paris will begin using artificial intelligence and facial analysis to determine whether students are paying attention in class. The school says the technology will be used to improve performance of students and professors.

Source: The Verge

Company offers free training for coal miners to become wind farmers

A Chinese wind-turbine maker wants American workers to retrain and become wind farmers. The training program was announced at an energy conference in Wyoming, where the American arm of Goldwind, a Chinese wind-turbine manufacturer is located.

Source: Quartz

Google AI defeats human Go champion

Google's DeepMind AI AlphaGo has defeated the world's number one Go player Ke Jie. AlphaGo secured the victory after winning the second game in a three-part match. DeepMind founder Demis Hassabis said Ke Jie "pushed AlphaGo right to the limit".

Source: BBC

Vegan burgers that taste like real meat to hit Safeway stores

Beyond Meat, which promises its plant-based burgers bleed and sizzle like real ground beef and is backed by investors like Bill Gates, will begin distributing its plant-based burgers in more than 280 Safeway stores in California, Hawaii and Nevada.

Source: Bloomberg

The brain starts to eat itself after chronic sleep deprivation

Brain cells that destroy and digest worn-out cells and debris go into overdrive in mice that are chronically sleep-deprived. The discovery could explain why a chronic lack of sleep puts people at risk of neurological disorders like Alzheimer’s disease.

Source: New Scientist

"We can still act and it won’t be too late," says Obama

Former US President Barack Obama has written an op-ed piece in the Guardian giving his views on some of the greatest challenges facing the world – food and climate change – and what we can do about them. "We can still act and it won’t be too late," writes Obama.

Source: The Guardian

Juno mission: Jupiter’s magnetic field is even weirder than expected

It has long been known that Jupiter has the most intense magnetic field in the solar system, but the first round of results from NASA’s Juno mission has revealed that it is far stronger and more misshapen than scientists predicted.

Announcing the findings of the spacecraft’s first data-collection pass, which saw Juno fly within 2,600 miles (4,200km) of Jupiter on 27th August 2016, NASA mission scientists revealed that the planet far surpassed the expectations of models.

Measuring Jupiter’s magnetosphere using Juno’s magnetometer investigation (MAG) tool, they found that the planet’s magnetic field is even stronger than models predicted, at 7.766 Gaus: 10 times stronger than the strongest fields on Earth.

Furthermore, it is far more irregular in shape, prompting a re-think about how it could be generated.

“Juno is giving us a view of the magnetic field close to Jupiter that we’ve never had before,” said Jack Connerney, Juno deputy principal investigator and magnetic field investigation lead at NASA’s Goddard Space Flight Center in Greenbelt, Maryland.

“Already we see that the magnetic field looks lumpy: it is stronger in some places and weaker in others.

An enhanced colour view of Jupiter’s south pole. Image courtesy of NASA/JPL-Caltech/SwRI/MSSS/Gabriel Fiset. Featured image courtesy of NASA/SWRI/MSSS/Gerald Eichstädt/Seán Doran

At present, scientists cannot say for certain why or how Jupiter’s magnetic field is so peculiar, but they do already have a theory: that the field is not generated from the planet’s core, but in a layer closer to its surface.

“This uneven distribution suggests that the field might be generated by dynamo action closer to the surface, above the layer of metallic hydrogen,” said Connerney.

However, with many more flybys planned, the scientists will considerable opportunities to learn more about this phenomenon, and more accurately pinpoint the bizarre magnetic field’s cause.

“Every flyby we execute gets us closer to determining where and how Jupiter’s dynamo works,” added Connerney.

With each flyby, which occurs every 53 days, the scientists are treated to a 6MB haul of newly collected information, which takes around 1.5 days to transfer back to Earth.

“Every 53 days, we go screaming by Jupiter, get doused by a fire hose of Jovian science, and there is always something new,” said Scott Bolton, Juno principal investigator from the Southwest Research Institute in San Antonio.

A newly released image of Jupiter’s stormy south pole. Image courtesy of NASA/JPL-Caltech/SwRI/MSSS/Betsy Asher Hall/Gervasio Robles

An unexpected magnetic field was not the only surprise from the first data haul. The mission also provided a first-look at Jupiter’s poles, which are unexpectedly covered in swirling, densely clustered storms the size of Earth.

“We’re puzzled as to how they could be formed, how stable the configuration is, and why Jupiter’s north pole doesn’t look like the south pole,” said Bolton. “We’re questioning whether this is a dynamic system, and are we seeing just one stage, and over the next year, we’re going to watch it disappear, or is this a stable configuration and these storms are circulating around one another?”

Juno’s Microwave Radiometer (MWR) also threw up some surprises, with some of the planet’s belts appearing to penetrate down to its surface, while others seem to evolve into other structures. It’s a curious phenomenon, and one which the scientists hope to better explore on future flybys.

“On our next flyby on July 11, we will fly directly over one of the most iconic features in the entire solar system – one that every school kid knows – Jupiter’s Great Red Spot,” said Bolton.

“If anybody is going to get to the bottom of what is going on below those mammoth swirling crimson cloud tops, it’s Juno and her cloud-piercing science instruments.”