SMART: Scalable Mesh-free Aerodynamic Simulations from Raw Geometries using a Transformer-based Surrogate Model

2026-01-26

AI Summary

Traditional grid-based physical simulations incur high computational costs when generating meshes for new geometries, while existing mesh-free methods struggle to balance accuracy and efficiency. This work proposes a Transformer-based mesh-free surrogate model that, given only a geometry point cloud and simulation parameters, accurately predicts physical fields at arbitrary locations. The method achieves—without requiring any simulation mesh—a level of accuracy that matches or even exceeds that of mesh-dependent approaches, enabling industrial-scale aerodynamic simulations. By integrating point cloud encoding, a shared latent space representation, a physics-informed decoder, and cross-layer attention mechanisms, the model demonstrates superior accuracy and efficiency across multiple complex geometries, outperforming or rivaling state-of-the-art grid-based methods.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Low Level & Physics-based Vision
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Machine learning-based surrogate models have emerged as more efficient alternatives to numerical solvers for physical simulations over complex geometries, such as car bodies. Many existing models incorporate the simulation mesh as an additional input, thereby reducing prediction errors. However, generating a simulation mesh for new geometries is computationally costly. In contrast, mesh-free methods, which do not rely on the simulation mesh, typically incur higher errors. Motivated by these considerations, we introduce SMART, a neural surrogate model that predicts physical quantities at arbitrary query locations using only a point-cloud representation of the geometry, without requiring access to the simulation mesh. The geometry and simulation parameters are encoded into a shared latent space that captures both structural and parametric characteristics of the physical field. A physics decoder then attends to the encoder's intermediate latent representations to map spatial queries to physical quantities. Through this cross-layer interaction, the model jointly updates latent geometric features and the evolving physical field. Extensive experiments show that SMART is competitive with and often outperforms existing methods that rely on the simulation mesh as input, demonstrating its capabilities for industry-level simulations.

Problem

• mesh-free simulation
• aerodynamic simulation
• surrogate model
• point-cloud geometry
• computational efficiency

Innovation

• mesh-free
• transformer-based surrogate model
• point-cloud geometry
• physics-informed decoding
• scalable aerodynamic simulation