China’s ‘mind-blowing’ claims – New AI system can design hypersonic weapons on its own without human intervention


Although the United States remains the leader in the field of anti-submarine warfare and artificial intelligence, China’s efforts to exploit “quality data” for military purposes have put Beijing far ahead of the world. Pentagon, said a former US Navy admiral.

A new paper published in a journal run by China’s aerospace defense industry claims that Beijing has made significant progress in building an AI (artificial intelligence) system capable of autonomously designing new hypersonic weapons.

“It was frustrating for all of us to work with large Excel spreadsheets, deciding which programs to push forward and which to hold back,” Admiral Bill Moran (retired), former vice chief of US naval operations, said during of a Navy League webinar. on artificial intelligence in November 2021.

On March 16, a team of researchers led by Professor Le Jialing from the China Aerodynamics Research and Development Center in Mianyang, Sichuan, published their findings in the Propulsion Technology Journal.

Le has been advising the Chinese military on hypersonic weapons technology for more than three decades, according to publicly available information.

As hypersonic research in China advances to Mach 8 – eight times the speed of sound – and beyond, the volume of experimental data to be processed and analyzed has also increased significantly, the researchers say.

In such a situation, the human brain can no longer keep up with the rapid pace of development of hypersonic technology according to Le and his team.

Countries around the world are racing to achieve hypersonic flight capability and an essential part of this race are simulation experiments that virtually create extreme hypersonic flight conditions in “wind tunnels”.

File Image: China’s hypersonic wind tunnel

How AI Systems Identify “Shockwaves”

When a missile or airborne object approaches at speeds greater than that of sound, it experiences what is called a “shock wave”, which is essentially a disturbance in the air around the vehicle that can cause changes extremely violent pressure on its surface.

So, scientists use wind tunnels to blow high-velocity wind onto their vehicle designs to see if their designs can perform desirably under such conditions.

Chinese hypersonic aircraft
Image of what is believed to be a Chinese hypersonic aircraft. (CCTV screenshot)

Now, there are different types of shock waves that can have different impacts on the vehicle and distinctly identifying these types of shock waves is essential for the design of hypersonic vehicles.

Each wind tunnel experiment can produce tens and thousands of simulated images of atmospheric disturbances around the vehicle and these photos must be manually studied by experienced researchers, often pixel by pixel, to identify which disturbance is a shock wave or what type shock wave it is.

An image from a hypersonic wind tunnel test contains a large amount of turbulence, and it may take human experts “an enormous amount of time and energy to label critical shock wave structures pixel by pixel,” said Le and his colleagues in the article.

Li’s team claims to have built an artificial intelligence machine that can identify most shock waves occurring during wind tunnel testing without even being told what to look for.

Usually, AI systems have to be taught by humans in a typical training session which would involve researchers carefully describing a shockwave tagging it with information so that the AI ​​can then identify it on its own. same.

The AI ​​will make mistakes initially, so it will need to be corrected over and over and this is how it will learn to correctly identify a particular shockwave. However, Li claims that his AI system needed no training!

The researchers used a technique called “unsupervised segmentation” based on a mathematical theory about graphs that can form a relationship between seemingly unrelated objects.

The machine would label what it believed to be a shockwave by examining the location, brightness, and color of each pixel. The AI ​​would use these early results as training material to continually improve its performance in recognizing shockwaves until it could detect shockwave patterns.

DF-17
Chinese DF-17 hypersonic missiles. (Image: China Military Online)

perfect match

According to the researchers, the shockwaves identified by their AI matched 85% of those marked by human experts.

Moreover, the overall accuracy of the AI ​​system was nearly 4 times higher than that of traditional computer software and it was based on a 3-year-old low-cost graphics card that took around 9 seconds to process an image.

If true, such a remarkable feat may give China an advantage not only in hypersonic vehicle design, but also in military applications such as autonomous target detection and learning-based weapon recognition. depth and deep neural networks.

“Our future cruise missiles will have a very high level of AI and autonomy,” so commanders can “control them in real time, or use a fire-and-forget mode, or even add more.” in-flight missile tasks,” a Chinese missile designer said in 2016.

Last year, PLA missile scientists at Rocket Force Engineering University said the accuracy of hypersonic weapons could be improved more than 10 times if full control was given to the machine.

They had published a paper detailing how the AI ​​can write the weapon’s software “on the fly”, as it travels at hyperspeed, thanks to a unique flight control algorithm.

Previous Inspiring Women: Speakers Announced for Free Webinar on Maximizing the Benefits of Mentoring
Next Innovative AI technology helps personalized ca