China touts an AI that can design its own hypersonic weapons

Although the US remains the leader in the anti-submarine warfare and Artificial Intelligence (AI)  China’s efforts to harness “quality data” for military use have put the country way ahead of the Pentagon, a former admiral of the US Navy said. And now, on the back of Creative AI’s already innovating and designing everything from new alloys, computer chips and drugs, to nerve agents and rocket engines, a new paper published in a journal run by China’s aerospace defense industry claims that Beijing has made significant progress in developing an AI that can design new hypersonic weapons autonomously.

As hypersonic weapons research in China comfortably passes Mach 8, and even Mach 20 and above thanks to a new wind tunnel capable of testing vehicles up to the crazy speed of Mach 30, the volume of experimental data that needs to be analysed and processed is increasing exponentially says Professor Le Jialing with China Aerodynamics Research and Development Centre in Mianyang, Sichuan. As a result humans it seems can no longer keep up so as countries around the world chase “hypersonic superiority” so the Chinese are turning to AI to help them not only analyse the data but to help them design new hypersonic weapons systems as well.

An essential part of this race is the use of simulations that virtually re-create the extreme hypersonic flight conditions found in wind tunnels, and when a missile or an aerial object approaches speeds exceeding that of sound it experiences a shock wave, which is basically a disturbance in the air around the vehicle that can cause extremely violent changes in pressure over its surface.

Unsurprisingly each wind tunnel experiment can produce tens of thousands of simulated images of atmospheric disturbances around the hypersonic vehicle which have to be studied manually by experienced researchers, often pixel by pixel, in order to identify which disturbance is a shock wave or what type of shock wave it is, and this process takes human experts “an enormous amount of time and energy to label the critical shock wave structures pixel by pixel” says Le.

Now though Le’s team claim to have built an AI that can identify most of these shock waves without even being instructed on what to look for. Usually, the AI systems need to be taught by humans in a typical training session which would involve the researchers carefully outlining a shock wave by labelling it with information so that the AI can then go on to identify them itself, and sometimes the AI will make mistakes which then have to be corrected. However, Le claims that his new AI system needs no training at all.

To do this they used a technique called unsupervised segmentation based on a mathematical theory on graphics that can form a relationship between seemingly unrelated objects. The machine labels what it believes to be a shock wave by examining the location, brightness, and colour of each pixel then uses these outputs as training material to continuously improve its performance in recognising shock waves until it can detect them all accurately and without any assistance.

According to the researchers, the shock waves identified by their AI were an 85 percent match to those marked by human experts. Furthermore, the AI system’s overall accuracy was nearly 4 times that of traditional computer software and ran on a low cost 3 year old graphic card that took about 9 seconds to process each image.

If true, then this could give China yet another edge in the race to build the best hypersonic weapons systems and missiles, as well as defense systems, and would again leave the US and others playing catch up in what’s now a very competitive race.

“Our future cruise missiles will have a very high level of AI and autonomy,” such that commanders will be able “to control them in a real-time manner, or to use a fire-and-forget mode, or even to add more tasks to in-flight missiles,” a senior Chinese missile designer said in 2016. And last year, PLA missile scientists from Rocket Force Engineering University had said that the accuracy of hypersonic weapons could be improved by more than 10 times if complete control is given to the machine.