Recent advances in neural surrogate modeling offer the potential for transformative innovations in applications such as automotive aerodynamics. Yet, industrial-scale problems often involve volumetric meshes with cell counts reaching 100 million, presenting major scalability challenges. Complex geometries further complicate modeling through intricate surface-volume interactions, while quantities such as vorticity are highly nonlinear and must satisfy strict divergence-free constraints. To address these requirements, we introduce AB-UPT as a novel modeling scheme for building neural surrogates for CFD simulations. AB-UPT is designed to: (i) decouple geometry encoding and prediction tasks via multi-branch operators; (ii) enable scalability to high-resolution outputs via neural simulation in a low-dimensional latent space, coupled with anchored neural field decoders to predict high-fidelity outputs; (iii) enforce physics consistency by a divergence-free formulation. We show that AB-UPT yields state-of-the-art predictive accuracy of surface and volume fields on automotive CFD simulations ranging from 33 thousand up to 150 million mesh cells. Furthermore, our anchored neural field architecture enables the enforcement of hard physical constraints on the physics predictions without degradation in performance, exemplified by modeling divergence-free vorticity fields. Notably, the proposed models can be trained on a single GPU in less than a day and predict industry-standard surface and volume fields within seconds. Additionally, we show that the flexible design of our method enables neural simulation from a CAD geometry alone, thereby eliminating the need for costly CFD meshing procedures for inference.
DrivAerStar: An Industrial-Grade CFD Dataset for Vehicle Aerodynamic Optimization
NVIDIA;Baidu, Inc.(百度公司);Peking University(北京大学);State Key Lab of General AI, Peking University(北京大学通用人工智能国家重点实验室);Beijing Key Laboratory of Behavior and Mental Health, Peking University(北京市行为与心理健康重点实验室);Embodied Intelligence Lab, PKU-Wuhan Institute for Artificial Intelligence(embodied intelligence lab, PKU-Wuhan人工智能研究所)
A Synergistic Framework Leveraging Autoencoders and Generative Adversarial Networks for the Synthesis of Computational Fluid Dynamics Results in Aerofoil Aerodynamics
TripNet: Learning Large-scale High-fidelity 3D Car Aerodynamics with Triplane Networks
Department of Mechanical Engineering(机械工程系);Massachusetts Institute of Technology(麻省理工学院);Department of Computer Science(计算机科学系);Technical University of Munich(慕尼黑技术大学)
CarBench: A Comprehensive Benchmark for Neural Surrogates on High-Fidelity 3D Car Aerodynamics
Massachusetts Institute of Technology(麻省理工学院);Schwarzman College of Computing, Massachusetts Institute of Technology(斯克里文计算机学院,麻省理工学院);Future Product Innovation, Toyota Research Institute(丰田研究机构未来产品创新部)
DrivAer Transformer: A high-precision and fast prediction method for vehicle aerodynamic drag coefficient based on the DrivAerNet++ dataset
Hubei Key Laboratory of Advanced Technology of Automotive Parts, Wuhan University of Technology, Wuhan, China(湖北汽车零部件先进技术重点实验室,武汉理工大学,中国武汉);Wuhan National Laboratory for Optoelectronic, Huazhong University of Science(武汉光电国家实验室,华中科技大学)
AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling
Universidad Politécnica de Madrid(马德里理工大学);University of Michigan(密歇根大学);Universitat Politècnica de València(瓦伦西亚理工大学);Purdue University(普渡大学);Caltech(加州理工学院);University of Washington(华盛顿大学)
Reinforcement Learning for Accelerated Aerodynamic Shape Optimisation
Software Competence Center Hagenberg (SCCH)(软件竞争力中心哈根贝格);Institut für Strömungsmechanik und Wärmeübertragung, TU Wien(流体力学与传热研究所,维也纳技术大学);CERBSim GmbH(CERBSim公司)
AI Agents in Engineering Design: A Multi-Agent Framework for Aesthetic and Aerodynamic Car Design
Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA(麻省理工学院机械工程系);Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA(麻省理工学院电气工程与计算机科学系);Department of Computer Science, Technical University of Munich, Munich, Germany(慕尼黑技术大学计算机科学系)
DrivAerNet: A Parametric Car Dataset for Data-Driven Aerodynamic Design and Prediction
Massachusetts Institute of Technology(麻省理工学院);Technical University of Munich(慕尼黑技术大学);Department of Mechanical Engineering(机械工程系);Department of Computer Science(计算机科学系)
DeepCFD: Efficient Steady-State Laminar Flow Approximation with Deep Convolutional Neural Networks
German Aerospace Center (DLR)(德国航空航天中心(DLR));German Research Center for AI (DFKI)(德国人工智能研究中心(DFKI));NUST School of Electrical Engineering and Computer Science(巴基斯坦国立科技大学)
Knowledge-Constrained Shape Optimization with a Mixture-of-Experts Neural Operator for High-Confidence Design
School of Ocean and Civil Engineering, Shanghai Jiao Tong University(上海交通大学海洋与土木工程学院);Eastern Institute of Technology(东方理工学院);TenFong Technology Co., Ltd.(十方科技有限公司);IM Motors Technology Co., Ltd.(集度汽车科技有限公司)