Lifeboat Foundation

Microcontrollers, miniature computers that can run simple commands, are the basis for billions of connected devices, from internet-of-things (IoT) devices to sensors in automobiles. But cheap, low-power microcontrollers have extremely limited memory and no operating system, making it challenging to train artificial intelligence models on “edge devices” that work independently from central computing resources.

Training a machine-learning model on an intelligent edge device allows it to adapt to new data and make better predictions. For instance, training a model on a smart keyboard could enable the keyboard to continually learn from the user’s writing. However, the training process requires so much memory that it is typically done using powerful computers at a data center, before the model is deployed on a device. This is more costly and raises privacy issues since user data must be sent to a central server.

To address this problem, researchers at MIT and the MIT-IBM Watson AI Lab have developed a new technique that enables on-device training using less than a quarter of a megabyte of memory. Other training solutions designed for connected devices can use more than 500 megabytes of memory, greatly exceeding the 256-kilobyte capacity of most microcontrollers (there are 1,024 kilobytes in one megabyte).

Although just cute little creatures at first glance, the microscopic geckos and octopuses fabricated by 3D laser printing in the molecular engineering labs at Heidelberg University could open up new opportunities in fields such as microrobotics or biomedicine.

The printed microstructures are made from novel materials —known as smart polymers—whose size and mechanical properties can be tuned on demand and with high precision. These “life-like” 3D microstructures were developed in the framework of the “3D Matter Made to Order” (3DMM2O) Cluster of Excellence, a collaboration between Ruperto Carola and the Karlsruhe Institute of Technology (KIT).

“Manufacturing programmable materials whose mechanical properties can be adapted on demand is highly desired for many applications,” states Junior Professor Dr. Eva Blasco, group leader at the Institute of Organic Chemistry and the Institute for Molecular Systems Engineering and Advanced Materials of Heidelberg University.

Cassie, developed at Oregon State University, now has a place in the “Guinness Book of World Records” for the fastest 100 meters by a bipedal robot.

The company targets a price of $3 per passenger per mile.

The first self-flying, all-electric, four-passenger eVTOL air taxi in the world was unveiled by the California-based Advanced Air Mobility (AAM) company Wisk Aero. Generation 6 is Wisk’s go-to-market aircraft and the first autonomous eVTOL to be a candidate for type approval by the Federal Aviation Administration (FAA).

Continue reading “Wisk Aero aims to be the first FAA certified autonomous air taxi with its 6th generation eVTOL” »

© All Right Reserved Science World 2022

Theme Trend News By WP News Theme

Tesla’s AI Day 2022 was mainly a recruiting event, according to CEO Elon Musk.

For Immediate Release.

IDTechEx Asks if Tesla’s Humanoid Robot “Optimus” Will Really Be Able to Serve People

IDTechEx.

I’m not a robotics expert, so I’ve been particularly keen to hear what robotics experts think of Tesla’s Optimus presentation the other day. The core arguments from Elon Musk and many Tesla fans regarding why Optimus is such a big deal are: Tesla will find a way to mass produce it at relatively low cost, Tesla is adding a brain to the robot, and it needs to be in the form of a human so that it can perform tasks designed to be done by humans. I don’t see any strong arguments against those things, but I know they are broad-brushed claims and quite vague. What about the details that I can’t see, that a common Tesla fan can’t see, and that perhaps even an engineer working on Optimus can’t see?

Let’s start with Dennis Hong. Dennis is a professor of mechanical & aerospace engineering at UCLA. He’s Director of RoMeLa: Robotics & Mechanisms Laboratory. With this title and being an independent expert in the separate world of academia, I was particularly interested to see his opinion. He was clearly excited as AI Day 2 arrived, but not in a sycophantic way. Luckily, he put his thoughts in a good little 13-post Twitter thread.

Scientists trained a machine learning tool to capture the physics of electrons moving on a lattice using far fewer equations than would typically be required, all without sacrificing accuracy. A daunting quantum problem that until now required 100,000 equations has been compressed into a bite-size task of as few as four equations by physicists using artificial intelligence. All of this was accomplished without sacrificing accuracy. The work could revolutionize how scientists investigate systems containing many interacting electrons. Furthermore, if scalable to other problems, the approach could potentially aid in the design of materials with extremely valuable properties such as superconductivity or utility for clean energy generation.