AI could lower the cost of designing more efficient airplane wings
AI agents devised a way to reduce friction of an airplane wing model even without access to a costly computer simulation
Scientists have demonstrated that artificial intelligence could significantly reduce the cost of designing more efficient airplane wings. By training AI agents using simple computer simulations, researchers found a way to reduce friction on a virtual airplane wing model by 38 percent. This breakthrough could revolutionize the development of more efficient and sustainable vehicles across various industries.
The AI agents employed a trial-and-error approach known as reinforcement learning to adjust the fluid flow over virtual objects, such as wings, to minimize drag. Impressively, the AI agents successfully applied their learnings from simpler scenarios to more complex airplane wing models, achieving substantial reductions in friction.
The researchers hope their platform, HydroGym, will serve as a computational infrastructure that enables AI to become a widely-used tool for tackling fluid flow challenges in science and engineering. The potential impact of such advancements could be immense, with billions of dollars in fuel savings and significant reductions in greenhouse gas emissions.
Written by urgent.news from New Scientist's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
