AI & Robotics News and Discussions

[Yuli Ban]

[Yuli Ban]

[caltrek]

To Swim Like a Tuna, Robotic Fish Need to Change How Stiff Their Tails are in Real Time
by Daniel Quinn

https://theconversation.com/to-swim-lik ... ime-168046

Introduction:

(The Conversation) Underwater vehicles haven’t changed much since the submarines of World War II. They’re rigid, fairly boxy and use propellers to move. And whether they are large manned vessels or small robots, most underwater vehicles have one cruising speed where they are most energy efficient.

Fish take a very different approach to moving through water: Their bodies and fins are very flexible, and this flexibility allows them to interact with water more efficiently than rigid machines. Researchers have been designing and building flexible fishlike robots for years, but they still trail far behind real fish in terms of efficiency.

What’s missing?

I am an engineer and study fluid dynamics. My labmates and I wondered if something in particular about the flexibility of fish tails allows fish to be so fast and efficient in the water. So, we created a model and built a robot to study the effect of stiffness on swimming efficiency. We found fish swim so efficiently over a wide range of speeds because they can change how rigid or flexible their tails are in real time.

[Ozzie guy]

The 7 Biggest Artificial Intelligence (AI) Trends In 2022 according to Forbes.

https://www.forbes.com/sites/bernardmar ... 824ced2015

[weatheriscool]

Scientists Built an Artificial Intelligence to Finish Beethoven’s Tenth Symphony
https://www.extremetech.com/extreme/327 ... h-symphony
By Jessica Hall on September 30, 2021 at 10:44 am

Nobody knows what Beethoven had on his mind when he died, nor the plans he had for his unfinished work. There’s been a lot of speculation, but the arrow of time flies in only one direction, and it’s tough to pick the brain of a dead guy. That is, unless you plug all his work into an AI to figure out his style. Music historians, composers, and computer scientists have collaborated to produce a “finished version” of composer Ludwig von Beethoven’s unfinished 10th Symphony. The first public performance of the piece is scheduled for October 9.

Beethoven is widely known to have intended to write a 10th Symphony, but he died with the work scarcely begun. Discussions of what he might have written continue into the present day. Up until now, the 10th Symphony was known only from fragments. A musicologist named Barry Cooper assembled Symphony No. 10’s first movement from these fragments back in the late 1980s, but this new project went farther and attempted to complete the second, third, and fourth movements using AI tools

[Yuli Ban]

[Nanotechandmorefuture]

I suppose if the AI is decent enough it would make sense things are improving on the robotics side as well. I guess we'll see how much improvement that will be over time. Personally over here in the USA once we start seeing videos of robots on the shooting range with 100% accuracy it won't be long until Officer Friendly is a robot. I would imagine by then society would be unrecognizable compared to years past so who knows what new things will happen once robots can walk on their own.

[caltrek]

Back to robotics...

Dyson’s Next Robot Vacuum Cleaner Revealed in FCC Filings With New Design
by James Vincent
October 4, 2021

https://www.theverge.com/circuitbreaker ... fcc-filing

Introduction:

(The Verge) Dyson’s last vacuum cleaner, the oddly-named 360 Heurist, skipped a US release, apparently because its design was ill-suited to the layout of American homes. But it looks like the company is planning to launch a new robovac in the US soon, after filing details of an unannounced product with the FCC. Photos in the filing seem to show a new design that is wider and flatter than its predecessors, with a squared-off rear and bulkier charging base.

The unnamed robovac comes in the same striking nickel blue color as its predecessor, and has tank treads and a detachable dust bin, just like Dyson’s previous forays into this product category. It also looks like the device (labelled RB03) has the Heurist’s 360-degree camera with LED lighting ring. Those LEDs are important, as they ensure the device can navigate in the dark — handy if you want your robovac to clean when you’re out or asleep.

Branding on the device suggests it will also use Dyson’s new Hyperdymium motors. This is simply Dyson’s name for its latest brushless electric motors, which have appeared in its most recent stick vacuums. It also looks like you can remove part of the internal tubing (picture 4 below) that connects the bin to the brushes. That would be a neat upgrade, as the 360 Heurist had a tendency to clog when vacuuming piles of large debris.

The most interesting change for this product, though, is the new shape. Dyson’s previous vacuums, the 360 Eye and 360 Heurist, have been taller and narrower than rival products built by iRobot and the like. This allows them tighter cornering around items like chair legs, but stops them from vacuuming under some furniture. It’s hard to gauge the exact dimensions of this new product (despite the inclusion of rulers in pictures, the perspective is confusing), but it does seem to be wider and flatter than the Eye and Heurist. This would tally with Dyson’s past comments that the Heurist was not quite right for US homes. Allowing the vacuum to get under more items of furniture would certainly help with sales.

[weatheriscool]

A framework to enhance deep learning using first-spike times
https://techxplore.com/news/2021-10-fra ... spike.html
by Ingrid Fadelli , Tech Xplore

Researchers at Heidelberg University and University of Bern have recently devised a technique to achieve fast and energy-efficient computing using spiking neuromorphic substrates. This strategy, introduced in a paper published in Nature Machine Intelligence, is a rigorous adaptation of a time-to-first-spike (TTFS) coding scheme, together with a corresponding learning rule implemented on certain networks of artificial neurons. TTFS is a time-coding approach, in which the activity of neurons is inversely proportional to their firing delay.

"A few years ago, I started my Master's thesis in the Electronic Vision(s) group in Heidelberg," Julian Goeltz, one of the leading researchers working on the study, told TechXplore. "The neuromorphic BrainScaleS system developed there promised to be an intriguing substrate for brain-like computation, given how its neuron and synapse circuits mimic the dynamics of neurons and synapses in the brain."

When Goeltz started studying in Heidelberg, deep-learning models for spiking networks were still relatively unexplored and existing approaches did not use spike-based communication between neurons very effectively. In 2017, Hesham Mostafa, a researcher at University of California—San Diego, introduced the idea that the timing of individual neuronal spikes could be used for information processing. However, the neuronal dynamics he outlined in his paper were still quite different from biological ones and thus were not applicable to brain-inspired neuromorphic hardware.

"We therefore needed to come up with a hardware-compatible variant of error backpropagation, the algorithm underlying the modern AI revolution, for single spike times," Goeltz explained. "The difficulty lay in the rather complicated relationship between synaptic inputs and outputs of spiking neurons."

[weatheriscool]

LEONARDO, the bipedal robot, can ride a skateboard and walk a slackline

by Robert Perkins, California Institute of Technology
https://techxplore.com/news/2021-10-leo ... kline.html

Researchers at Caltech have built a bipedal robot that combines walking with flying to create a new type of locomotion, making it exceptionally nimble and capable of complex movements.

Part walking robot, part flying drone, the newly developed LEONARDO (short for LEgs ONboARD drOne, or LEO for short) can walk a slackline, hop, and even ride a skateboard. Developed by a team at Caltech's Center for Autonomous Systems and Technologies (CAST), LEO is the first robot that uses multi-joint legs and propeller-based thrusters to achieve a fine degree of control over its balance.

A paper about the LEO robot was published online on October 6 and was featured on the October 2021 cover of Science Robotics.

"We drew inspiration from nature. Think about the way birds are able to flap and hop to navigate telephone lines," says Soon-Jo Chung, corresponding author and Bren Professor of Aerospace and Control and Dynamical Systems. "A complex yet intriguing behavior happens as birds move between walking and flying. We wanted to understand and learn from that."

[Yuli Ban]

In the middle of the French lockdown in October 2020, a new restaurant launched on Deliveroo and UberEats in Paris: Cala.

Its pasta dishes started at €8 each, causing some customers to wonder about the high quality to price ratio — particularly compared to some other Parisian outlets.

It wasn’t until the lockdown ended and Cala customers could visit the restaurant in person that the reason became clear. Instead of a team of chefs, the food is cooked and assembled by a robot.

“We wanted to make sure that the quality of the product was what was really driving customers to come to a restaurant,” says Ylan Richard, who founded Cala in 2019, when he was 19 . “No one knew there was a robot behind the restaurant on the platforms.”

Cala’s now raised €5.5m in seed funding, led by Backed VC, and is scoring in the top 1% of Paris’ restaurants on delivery apps.

Cala is part of a new wave of food robots (some people say it’s spurred on by a pandemic-induced hygiene desire to remove humans from the food prep process). In Europe, there’s Pazzi, also in Paris, where a robot cooks pizzas; in Sweden, there’s ice cream robot BonBot and in the UK, there’s Karakuri.

It’s tech that could accelerate parts of the restaurant industry — particularly dark kitchens.

[Yuli Ban]

New robots patrolling for 'anti-social behaviour' causing unease in Singapore streets

There are new sheriffs in town in Singapore, and they are unnerving many who live there.

Called "Xavier," the robots are equipped with seven cameras that enable them to detect "undesirable social behaviour," for instance if you incorrectly park your bike, if you smoke in an unauthorised area or if social distancing is not being respected.

According to project manager Michael Lim, these machines are a new weapon against insecurity.

"If the robot is around and something happens, the people in the control room will have a trace and will be able to see what happened”, he said.

The robots initially patrolled a housing estate and a shopping centre as part of a 3-week trial in September.

[wjfox]

Robots Take Over Italy’s Vineyards as Wineries Struggle With Covid-19 Worker Shortages

Oct. 3, 2021 7:34 am ET

The coronavirus pandemic is pushing the wine industry toward automation.

Covid-related travel restrictions left severe shortages of agricultural workers last year, as Eastern Europeans and North Africans were unable to reach fields in Western Europe. Though the shortages have eased this year, the difficulty of finding workers has accelerated the shift, which was already under way across the agricultural sector.

While harvests of some crops, like soybeans and corn, are already heavily automated, winemakers have been slower to make the switch.

[...]

For many vintners in Europe and the U.S., however, the difficulty of finding workers—a problem they say had grown steadily for years but became acute during the pandemic—has pushed them to take the robot plunge. It is a change that will outlast the pandemic and could shift longstanding migration patterns that bring tens of thousands of foreign workers to Italy, France and Spain for agricultural harvests each year.

[...]

Last year was the worst labor shortage of his half-century career in wine. Use of harvesting machines among the group’s members increased 20% this year in response, he said.

https://www.wsj.com/articles/robots-tak ... 1633260841

[Yuli Ban]

We are excited to introduce the DeepSpeed- and Megatron-powered Megatron-Turing Natural Language Generation model (MT-NLG), the largest and the most powerful monolithic transformer language model trained to date, with 530 billion parameters. It is the result of a joint effort between Microsoft and NVIDIA to advance the state of the art in AI for natural language generation.
As the successor to Turing NLG 17B and Megatron-LM, MT-NLG has 3x the number of parameters compared to the existing largest model of this type and demonstrates unmatched accuracy in a broad set of natural language tasks such as:

• Completion prediction
• Reading comprehension
• Commonsense reasoning
• Natural language inferences
• Word sense disambiguation

The 105-layer, transformer-based MT-NLG improved upon the prior state-of-the-art models in zero-, one-, and few-shot settings and set the new standard for large-scale language models in both model scale and quality.

[andmar74]

[raklian]

Yep, we are living in the future. Dystopia time!

[Nanotechandmorefuture]

Yep, we are living in the future. Dystopia time!

That's for 1st generation. I bet the later ones won't be so easy and besides with that robot dog with the gun that has been going around recently good luck with the later iterations.

[Yuli Ban]

My gosh, look at this!


It looks so biological....

[Yuli Ban]

[Yuli Ban]