Photo credit: venturebeat.com
Rune Aero has transformed its aircraft development process by utilizing an interactive virtual wind tunnel that employs Physics AI technology. This innovative approach has led to an impressive 80% reduction in preliminary design costs, while simultaneously doubling the payload capacity and halving fuel consumption.
The cutting-edge digital twin technology integrates Luminary Cloud’s Nvidia CUDA-X-enhanced CFD simulations with an Nvidia Omniverse Blueprint for real-time computer-aided engineering digital twins. This technology allows engineers to instantly see the aerodynamic effects as design modifications are made. The result is a streamlined optimization process that replaces traditional iterative physical testing methods, which could take months, with ongoing real-time adjustments, particularly advantageous for the company focusing on autonomous cargo aircraft.
The official announcement was made during Nvidia’s GTC 2025 conference held in San Jose, California.
AI Physics Transforms Aircraft Economics with Real-Time Design Optimization
By engaging in virtual wind tunnel testing during the early design stages, Rune Aero significantly mitigates future technical risks and associated costs compared to exclusively relying on traditional physical wind tunnel evaluations.
The benefits reported by the company are significant: cargo operators could see a 70% cut in operating costs due to enhanced aerodynamics and fuel efficiency, coupled with doubled payload capacity and a 50% decrease in fuel consumption. For air cargo operators, these enhancements could lead to substantial profitability increases, fundamentally reshaping the economics of air freight.
Luminary Cloud played a key role in developing this interactive virtual wind tunnel by merging its GPU-native CFD solvers with Nvidia CUDA-X libraries, the PhysicsNeMo AI training framework, and Omniverse visualization APIs. This multifaceted technology enables the creation of digital twin environments that provide immediate feedback on design changes. Additionally, this framework allows for the integration of real-world data into the Physics AI platform, opening up further possibilities for digital twin applications.
Tim Costa, senior director CAE EDA and quantum at Nvidia, emphasized, “AI Physics models are crucial for real-time interactive engineering analysis and design, yet they demand vast amounts of data. Luminary Cloud’s accelerated CFD solver empowers clients to produce the high-fidelity data essential for training the underlying AI physics model in Nvidia PhysicsNeMo within hours.”
Juan J. Alonso, CTO of Luminary Cloud, remarked, “It’s thrilling to collaborate with aerospace ventures like Rune Aero that are prepared to implement innovative design optimization methods. By swiftly executing extensive simulations and developing Physics AI models, Rune Aero’s engineers can quickly arrive at viable designs to address critical regional air transport challenges.”
Nadine Auda, co-founder of Rune Aero, stated, “Fast, precise, and cost-efficient design feedback is vital during the initial phases of aircraft development. Luminary’s virtual wind tunnel equips us to test design configurations early, resulting in an over 80% reduction in early development costs compared to traditional wind tunnels, and enabling more informed, rapid design choices. The optimized aircraft configurations derived from these tests enhance the aerodynamic lift-to-drag ratio, and when combined with advanced propulsion systems, they double both payload and range while halving fuel consumption, ultimately lowering operational expenses for our clients.”
Moreover, Luminary Cloud and nTop have announced a new partnership with Nvidia PhysicsNeMo that significantly accelerates physics-based AI design optimization, reducing the timeframe from weeks or months to just hours. This integration streamlines the connection between nTop’s parametric geometry generation, Luminary’s GPU-native simulation platform, and Nvidia PhysicsNeMo through NIM microservices, enabling engineers to generate and evaluate hundreds of design variations within a single day—an impressive improvement over the previous lengthy manual process.
Source
venturebeat.com