Significant advances in AI in recent years have led to the growing sophistication of military weapons leading to the creation of Lethal Autonomous Weapon Systems (L.A.W.S)., These are machines engineered to identify, engage and destroy targets without human control. The Chinese researchers have made important progress in building an AI system that can design hypersonic weapons by itself. Countries across the world are racing to achieve hypersonic flight capability. The essential part of this race are simulation experiments, that can create extreme hypersonic flight conditions virtually, in the wind tunnels. 

The hypersonic research in China advances to Mach 8- eight times the speed of sound. The volume of experimental data to be processed and analyzed has also been increasing according to the researchers. In situations like these, human brains can no longer keep up with the rapid pace of hypersonic technology development. 

It was on March 16th, a team of researchers led by professor Le Jialing with the China Aerodynamics Research and Development Center in Mianyang Sichuan, published their findings in the Journal of Propulsion Technology. According to publicly available information about Le, he has been an advisor to the Chinese military on hypersonic weapon technology for more than three decades.  

 Identifying ‘Shock Waves' 

 When a missile or an aerial object approaches a speed exceeding that of sound, it experiences a phenomenon called the ‘shock wave’. This occurrence is a disturbance in the air around the vehicle. It can cause violent changes in pressure over its surface. Therefore, scientists use wind tunnels to blow the wind at high speeds over their vehicle designs. This is to see if the vehicle can perform desirably under such conditions. 

There are different types of shock waves that can have differing impacts on vehicles. It is very important to identify various kinds of shock waves for designing hypersonic vehicles. Each wind tunnel experiment can produce thousands of simulated images of atmospheric disturbances. The manual study of these photos has to be done, often pixel by pixel, to identify which disturbance is a shock wave or what type of wave is it. This manual analysis is a time-consuming process. This is where AI comes in. 

Le claims to have built an AI machine that could identify most of the shock waves without the requirement of human intervention. This AI-grounded machine could identify most of the shock waves occurring in wind tunnel tests without even being instructed on what to look for.  

In normal cases, the AI system needs to be trained by humans and there are chances that this system makes mistakes in the initial stages. In this case, according to a statement by Le, the machine needed no training at all. A technique called ‘unsupervised segmentation’ was used by the researchers based on a mathematical theory on graphics that can form a relationship between seemingly unrelated objects. Mentioned following are the factors that the machine takes into consideration while analyzing the wave: 

  1. Location 
  2. Brightness 
  3. Color of each pixel 

The AI would use the initial result as the training material to continuously improve its performance in identifying the shock waves until it detects the pattern. 

Initial Results  

According to the researchers, the shock waves identified by their model was an 85% match to those marked by the human experts. The system's overall accuracy was nearly 4 times that of traditional computer software. It was built on a low cost 3-year-old graphic card that will take only 9 seconds to process an image.  

This remarkable feat can give China an edge not only in hypersonic vehicles but also in military applications such as autonomous target detection and recognition of weapons. Last year PLA missile scientists stated that the accuracy of hypersonic weapons could be improved by 10 times if complete control was given to the machines.  

Sources of Article

Source: EurAsian Times

Want to publish your content?

Publish an article and share your insights to the world.

Get Published Icon
ALSO EXPLORE