In a breakthrough set to reconstruct the future of aviation combat, Chinese researchers claim that they have negated humanity’s ultimate tactical advantage over artificial intelligence (ai) Dogfight: the ability to overtake algorithms through unpredictable, high-strength acrobatics of aircraft.
The method, detailed in a study published later last year, combines advanced infrared imaging with AI-driven predictive modeling to predict opponent movements by detecting subtle wing tail movements.
This is a development that can effectively render even the most agile fighter jets, such as the US-made F-15s, virtually vulnerable. scientist From the Northwest Mechanical and Electrical Engineering Institute, based in the northwest city of Xianyang, the main research arm of Norinco, China’s largest weapons supplier.
The study, published in the December issue of Journal of Gun Launch & Control, addresses serious flaws in existing AI AIRs battle System: The reliance on orbit-based predictions struggles to explain sudden, nonlinear operations performed by human pilots.
The Chinese team, led by senior engineer Lin Zhiwei, circumvented this limitation by focusing on the physical mechanics of enemy aircraft.
Using the modified Yolov8 neural network, the system analyzes infrared images to detect millimeter-level deformations of opponent control surfaces, such as the 1.5m (5ft) ladder of the F-15 or 2m (6.5ft) elevator.
