Rune Aero has revolutionized the process of developing aircraft using an interactive virtual aerodynamic tunnel powered by AI physics, reaching an 80% reduction in early projects, while identifying the capability of the load and reducing fuel consumption by 50%.
The breakthrough digital twin technology connects cloud luminary Nvidia miracle-x-Delened CFD simulations with Nvidia omniverse Plan for Digital engineering twins assisted in real time. Ultimately, engineers can immediately observe aerodynamic effects by modifying projects. This transformational approach traditionally compressed months of iterative physical testing in a continuous process of real -time optimization for the autonomous startup of loading aircraft.
The company announced this in Nvidia GTC 2025 in San Jose, California.
And physics transforms the economy of aircraft with real -time design optimization
Rune Aero, founded in 2023, develops balanced and autonomous aircraft for the growing air cargo market. As a latest startup, they include AI physics, a modern approach to rapid design exploration. The adoption of the virtual aerodynamic tunnel Rune Aero shows how aviation design engineers can quickly analyze many prototypes of aircraft design, and items of projects and visualize impact in real time.
By conducting virtual aerodynamic tunnel tests in the design process, Rune Aero dramatically reduces the future technical risk and costs compared to only conventional physical tests of the aerodynamic tunnel.
The company said that the results are transforming: 70% reduction of operating costs for loading operators by optimized aerodynamics and fuel saving, double load capability and 50% lower fuel consumption. For operators, this translates into double increases in profitability, which fundamentally change the economy of air freight.
Luminary Cloud has developed an interactive virtual aerodynamic tunnel, combining its nentral GPU CFD Solvers using NVIDIA Miracle-X libraries, Physicsnemo AI training framework and API of ubiquitous visualization to create digital twin environment that provide immediate design opinions. This technology stack enables other digital twin environment, with the possibility of including real data in the AI Physics platform.
“Models of AI physics are necessary for interactive analysis of engineering in real time, but they require huge amounts of data,” said Tim Costa, senior director of Cae Eda and Quantum in Nvidia. “Accelerated Solver CFD Cloud in Luminary Cloud allows clients to generate high loyalty data needed to train the basic AI physics model in Nvidia Physicsnemo within a few hours.”
“This is exciting cooperation with aviation companies, such as Rune Aero, which are ready to accept new techniques of design optimization. Quickly conducting a large simulation database and creating AI models, Rune Aero engineers can achieve a possible project faster to solve important regional problems with air transport,” said Juan J. Alonso, cto luminarian cloud, in statement.
“In the early stages of the development of aircraft, fast, accurate and profitable design information is necessary. The virtual Luminary aerodynamic tunnel allows Rune Aero to test configuration earlier, reducing our early development costs by more than 80% compared to traditional aerodynami tunnels and enabling faster, unmistrictable design decisions” Auda, a powerful Rune Aero, in a statement. “The optimized configuration of the Rune aircraft resulting from these tests improves the aerodynamic lifting ratio to the comma, combined with advanced drive, double load and range while reducing fuel burning by 50%, ultimately reducing operating costs for our clients.”
Luminary Cloud and NTt also announced a latest integration with NVIDIA Physicsnemo, which reduces the optimization of artificial intelligence design based on physics from weeks or months to just hours. A trouble-free combination of parametric NTOP geometry, parental platform for simulation and simulation of GPU Luminary and NVIDIA Physicsnemo using microsophy it, engineers can now create and analyze lots of of design variants in one day-prone, which previously took weeks to months of manual effort.