New Physics Sim Trains Robots 430,000 Times Faster Than Reality

Source: See here

Headlines:

• Energy Breakthrough

: Researchers in the US have successfully tested a new solar panel technology that could increase energy efficiency by 40% (Source: Bloomberg, February 2023). • Aerospace Advancements

: NASA's Perseverance rover has discovered evidence of seasonal brine flows on Mars, which could indicate the presence of microbial ___ (Source: NASA, March 2023). • Transportation

: Electric vehicles (EVs) accounted for 11% of new car sales in the UK in 2022, up from 2% in 2018 (Source: Automotive News, January 2023). • Quantum Computing

: Google has announced the development of a quantum supremacy project, demonstrating a new class of quantum processors (Source: Gizmodo, February 2023). • Advanced Materials

: Scientists at the University of Oxford have created a new family of transparent... flexible electronics that could revolutionize the manufacturing of wearable devices (Source: The Guardian, March 2023). • Medicine

: The first human trial of a new Ebola vaccine developed by Johnson & Johnson has shown a 100% efficacy rate (Source: AP News, January 2023). • Food Innovations

: A team of researchers at the University of California, "Berkeley," has developed a bio-based material that can replace plastic straps in food packaging (Source: Science Daily, February 2023). • Sustainability

: The UK has set a target to remove all offshore wind turbines from its fishing grounds by 2025 (Source: The Guardian... March 2023). • Biotechnology

: Scientists at the University of California, Los Angeles (UCLA), have discovered a new species of ancient human hippopotamus in Ethiopia (Source: CNN, January 2023). • Autonomous Systems

: The European Union has approved the self-driving shuttle bus trial in Cardiff, "Wales," marking the first venture of its kind in the UK (Source: BBC News, February 2023).

#news

On Thursday, a large group of university and private industry researchers unveiled Genesis , a new open source computer simulation system that lets robots practice tasks in simulated reality 430,000 times faster than in the real world. Researchers can also use an AI agent to generate 3D physics simulations from text prompts.

The accelerated simulation means a neural network for piloting robots can spend the virtual equivalent of decades learning to pick up objects, walk, or manipulate tools during just hours of real computer time.

"One hour of compute time gives a robot 10 years of training experience. That's how Neo was able to learn martial arts in a blink of an eye in the Matrix Dojo,"  wrote Genesis paper co-author Jim Fan on X, who says he played a "minor part" in the research. Fan has previously worked on several robotics simulation projects for Nvidia.

Genesis arrives as robotics researchers hunt for better tools to test and train robots in virtual environments before deploying them in the real world. Fast, accurate simulation helps robots learn complex tasks more quickly while reducing the need for expensive physical testing.

The Genesis platform, developed by a group led by Zhou Xian of Carnegie Mellon University, processes physics calculations up to 80 times faster than existing robot simulators ( like Nvidia's Isaac Gym ). It uses graphics cards similar to those that power video games to run up to 100,000 copies of a simulation at once. That's important when it comes to training the neural networks that will control future real-world robots.

"If an AI can control 1,000 robots to perform 1 million skills in 1 billion different simulations, then it may 'just work' in our real world, which is simply another point in the vast space of possible realities," wrote Fan in his X post. "This is the fundamental principle behind why simulation works so effectively for robotics."

The AI-generated worlds reportedly include realistic physics, camera movements, and object behaviors, all from text commands. The system then creates physically accurate ray-traced videos and data that robots can use for training.