“Do Diffusion Models Dream of Electric Planes?” Discrete and Continuous Simulation-Based Inference for Aircraft Design

In this paper, we generate conceptual engineering designs of electric vertical take-off and landing (eVTOL) aircraft. We follow the paradigm of simulation-based inference (SBI), whereby we look to learn a posterior distribution over the full eVTOL design space. To learn this distribution, we sample over discrete aircraft configurations (topologies) and their corresponding set of continuous parameters. Therefore, we introduce a hierarchical probabilistic model consisting of two diffusion models. The first model leverages recent work on Riemannian Diffusion Language Modeling (RDLM) and Unified World Models (UWMs) to enable us to sample topologies from a discrete and continuous space. For the second model we introduce a masked diffusion approach to sample the corresponding parameters conditioned on the topology. Our approach rediscovers known trends and governing physical laws in aircraft design, while significantly accelerating design generation.

This paper shows how AI can help design new electric aircraft. We build a system that learns from thousands of computer‑simulated aircraft and can then propose entirely new designs that still obey physics. The key challenge is that aircraft differ not only in their shapes and sizes, but also in their basic layout, for example how many wings or rotors they have. To handle this, the system uses two diffusion models: one chooses the overall aircraft layout starting from desired goals, and the other fills in the continuous dimensions associated with the layout and those goals. The models are trained on more than 30,000 valid aircraft generated with an aerospace simulator. When asked for designs that meet certain goals, like carrying more weight, the system produces sensible configurations, such as adding extra wings. Overall, the work shows that modern generative models can explore large, complex engineering problems and recover meaningful aircraft design trends.