Today at AWS re:Invent 2024 in Las Vegas, London-based startup PhysicsX is launching the first large geometry model for aerospace engineering called LGM-Aero and a publicly accessible reference application called Ai.rplane to showcase its power in designing aerostructures. Ai.rplane allows engineers to generate innovative aircraft designs in an “infinitely wide” design space and instantaneously assess the designed aircraft’s potential performance.

“In the same way that large language models understand text, Ai.rplane has a vast knowledge of the shapes and structures that are important to aerospace engineering,” said Jacomo Corbo, Co-founder and CEO of PhysicsX. “The technology can optimize across multiple types of physics in seconds, many orders of magnitude faster than numerical simulation—and at the same level of accuracy. We’re excited about what LGM-Aero brings as capabilities to our customers while recognizing that it is also an important stepping stone towards developing physics foundation models.”

In one operation, the technology creates novel designs; predicts lift, drag, stability, structural stress, and other attributes for each shape; and then optimizes the design according to the user’s preferences. This workflow reduces development time from months to hours when used in industrial applications.

“This is a first step in transforming the way engineering is practiced in Advanced Industries,” added Robin Tuluie, Co-founder and Chairman of PhysicsX. “Over time, we will bring new capabilities to LGM-Aero and to Ai.rplane, allowing users to select powertrains, add controls, and further content to reach mature designs in days rather than months or years.”

LGM-Aero was developed and provisioned on AWS and trained on over 25 million meshes, representing over 10 billion vertices and tens of thousands of CFD (computational fluid dynamics) and FEA (finite element analysis) simulations generated with Siemens Digital Industries Xcelerator tools. The fully trained model generalizes to a broad set of aeroelastic applications and infers aero performance, flight stability, and structural stress for a large class of flying shapes as a zero-shot model.

“This technology will accelerate the transformation of engineering in advanced industries for AWS customers, enabling them to bring their products to the market faster while increasing product performance,” said Ozgur Tohumcu, General Manager of Automotive and Manufacturing at AWS.

LGM-Aero was developed using an extensive set of simulation technologies from Siemens to automate and scale the generation of high-quality training data, as well as AWS Batch and Amazon EC2 to scale compute during training. It is available on the PhysicsX AI engineering enterprise platform.

PhysicsX emerged from stealth in November 2023 to help apply the power of generative AI to enable breakthrough engineering in advanced industries including automotive, aerospace, renewables, and materials production. That month it announced it had raised $32 million in a Series A round led by General Catalyst, with Standard Investments, NGP, Radius Capital, and KKR Co-founder and Co-executive chairman, Henry Kravis, also participating.

Tuluie and Corbo, and fellow co-founder Nicolas Haag, Director of Simulation Engineering, are leading a PhysicsX team of simulation engineers, machine learning and software engineers, and data scientists with backgrounds spanning numerical physics, enterprise AI, and advanced engineering including some from Formula One. For the last three years, the team has worked with some of the largest and most sophisticated engineering and manufacturing organizations in the world on their most critical engineering challenges.

The company says that accelerating the energy transition hinges on overcoming fundamental engineering bottlenecks to the deployment of more efficient technologies and machines at scale, from time-consuming physics simulation to the painstaking reconciliation of virtual simulation and real-world data collection to the many limitations of optimizing over a large design space.

It aims to meet the needs of engineers across sectors by building AIs that dramatically accelerate accurate physics simulation, enable generative engineering solutions, vastly accelerate some of the most time-consuming activities in the engineering process, and make complex engineering across industries more accessible.