Abstract

This study introduces DrivAerNet, a large-scale high-fidelity CFD dataset of 3D industry-standard car shapes, and RegDGCNN, a dynamic graph convolutional neural network model for regression, both aimed at aerodynamic car design through machine learning. DrivAerNet, with its 4000 detailed 3D car meshes using 0.5 million surface mesh faces and comprehensive aerodynamic performance data comprising of full 3D pressure, velocity fields, and wall-shear stresses, addresses the critical need for extensive datasets to train deep learning models in engineering applications. It is 60% larger than the previously available largest public dataset of cars and is the only open-source dataset that also models wheels and underbody. RegDGCNN leverages this large-scale dataset to provide high-precision drag estimates directly from 3D meshes, bypassing traditional limitations such as the need for 2D image rendering or signed distance fields (SDFs). By enabling fast drag estimation in seconds, RegDGCNN facilitates rapid aerodynamic assessments, offering a substantial leap toward integrating data-driven methods in automotive design. Together, DrivAerNet and RegDGCNN promise to accelerate the car design process and contribute to the development of more efficient cars. To lay the groundwork for future innovations in the field, the dataset and code used in our study are publicly accessible.

References

1.
Brand
,
C.
,
Anable
,
J.
,
Ketsopoulou
,
I.
, and
Watson
,
J.
,
2020
, “
Road to Zero Or Road to Nowhere? Disrupting Transport and Energy in a Zero Carbon World
,”
Energy Policy
,
139
, p.
111334
.
2.
Martins
,
H.
,
Henriques
,
C.
,
Figueira
,
J.
,
Silva
,
C.
, and
Costa
,
A.
,
2023
, “
Assessing Policy Interventions to Stimulate the Transition of Electric Vehicle Technology in the European Union
,”
Socio-Econ. Plan. Sci.
,
87
, p.
101505
.
3.
Mock
,
P.
, and
Díaz
,
S.
,
2021
, “
Pathways to Decarbonization: The European Passenger Car Market in the Years 2021–2035
,”
Communications
,
49
, pp.
847129
848102
.
4.
Aultman
,
M.
,
Wang
,
Z.
,
Auza-Gutierrez
,
R.
, and
Duan
,
L.
,
2022
, “
Evaluation of CFD Methodologies for Prediction of Flows Around Simplified and Complex Automotive Models
,”
Comput. Fluids
,
236
, p.
105297
.
5.
Ayman
,
T.
,
Elrefaie
,
M. A.
,
Sayed
,
E.
,
Elrefaie
,
M.
,
Ayyad
,
M.
,
Hamada
,
A. A.
, and
Abdelrahman
,
M. M.
,
2023
, “
2023 5th Novel Intelligent and Leading Emerging Sciences Conference (NILES)
,”
NILES
,
Giza, Egypt
,
Oct. 21–23
, IEEE, pp.
157
160
.
6.
Elrefaie
,
M.
,
Ayman
,
T.
,
Elrefaie
,
M. A.
,
Sayed
,
E.
,
Ayyad
,
M.
, and
AbdelRahman
,
M. M.
,
2024
, “
Surrogate Modeling of the Aerodynamic Performance for Airfoils in Transonic Regime
,” AIAA SCITECH 2024 Forum, p.
2220
.
7.
Thuerey
,
N.
,
Weißenow
,
K.
,
Prantl
,
L.
, and
Hu
,
X.
,
2020
, “
Deep Learning Methods for Reynolds-Averaged Navier–Stokes Simulations of Airfoil Flows
,”
AIAA J.
,
58
(
1
), pp.
25
36
.
8.
Baque
,
P.
,
Remelli
,
E.
,
Fleuret
,
F.
, and
Fua
,
P.
,
2018
, “
Geodesic Convolutional Shape Optimization
,” Proceedings of the 35th International Conference on Machine Learning, Volume 80 of Proceedings of Machine Learning Research,
J.
Dy
A.
Krause
, eds,
PMLR
, pp.
472
481
.
9.
Song
,
B.
,
Yuan
,
C.
,
Permenter
,
F.
,
Arechiga
,
N.
, and
Ahmed
,
F.
,
2023
, “
International Design Engineering Technical Conferences and Computers and Information in Engineering Conference
,”
IDETC
,
Boston, MA
,
Aug. 20–23
.
10.
Li
,
Z.
,
Kovachki
,
N.
,
Choy
,
C.
,
Li
,
B.
,
Kossaifi
,
J.
,
Otta
,
S.
,
Nabian
,
M. A.
,
Stadler
,
M.
,
Hundt
,
C.
,
Azizzadenesheli
,
K.
, et al.,
2024
, “
Geometry-Informed Neural Operator for Large-Scale 3d PDEs
,”
Adv. Neural Inform. Process. Syst.
,
36
, pp.
35836
35854
.
11.
Heft
,
A. I.
,
Indinger
,
T.
, and
Adams
,
N. A.
,
2012
, “
Experimental and Numerical Investigation of the Drivaer Model
,” Fluids Engineering Division Summer Meeting, Vol. 44755,
American Society of Mechanical Engineers
, pp.
41
51
.
12.
Wang
,
Y.
,
Sun
,
Y.
,
Liu
,
Z.
,
Sarma
,
S. E.
,
Bronstein
,
M. M.
, and
Solomon
,
J. M.
,
2019
, “
Dynamic Graph CNN for Learning on Point Clouds
,”
ACM Trans. Graph.
,
38
(
5
), pp.
1
12
.
13.
Arechiga
,
N.
,
Permenter
,
F.
,
Song
,
B.
, and
Yuan
,
C.
,
2023
,
NeurIPS 2023 Workshop on Diffusion Models
.
14.
Gunpinar
,
E.
,
Coskun
,
U. C.
,
Ozsipahi
,
M.
, and
Gunpinar
,
S.
,
2019
, “
A Generative Design and Drag Coefficient Prediction System for Sedan Car Side Silhouettes Based on Computational Fluid Dynamics
,”
Comput. Aided Des.
,
111
, pp.
65
79
.
15.
Usama
,
M.
,
Arif
,
A.
,
Haris
,
F.
,
Khan
,
S.
,
Afaq
,
S. K.
, and
Rashid
,
S.
,
2021
, “
A Data-Driven Interactive System for Aerodynamic and User-Centred Generative Vehicle Design
,” 2021 International Conference on Artificial Intelligence (ICAI), pp.
119
127
.
16.
Bonnet
,
F.
,
Mazari
,
J.
,
Cinnella
,
P.
, and
Gallinari
,
P.
,
2022
, “
Airfrans: High Fidelity Computational Fluid Dynamics Dataset for Approximating Reynolds-Averaged Navier–Stokes Solutions
,”
Adv. Neural Inform. Process. Syst.
,
35
, pp.
23463
23478
.
17.
Kashefi
,
A.
, and
Mukerji
,
T.
,
2022
, “
Physics-Informed Pointnet: A Deep Learning Solver for Steady-State Incompressible Flows and Thermal Fields on Multiple Sets of Irregular Geometries
,”
J. Comput. Phys.
,
468
, p.
111510
.
18.
Umetani
,
N.
, and
Bickel
,
B.
,
2018
, “
Learning Three-Dimensional Flow for Interactive Aerodynamic Design
,”
ACM Trans. Graph.
,
37
.
19.
Remelli
,
E.
,
Lukoianov
,
A.
,
Richter
,
S.
,
Guillard
,
B.
,
Bagautdinov
,
T.
,
Baque
,
P.
, and
Fua
,
P.
,
2020
, “
Meshsdf: Differentiable Iso-surface Extraction
,”
Adv. Neural Inform. Process. Syst.
,
33
, pp.
22468
22478
.
20.
Rios
,
T.
,
van Stein
,
B.
,
Wollstadt
,
P.
,
Back
,
T.
,
Sendhoff
,
B.
, and
Menzel
,
S.
,
2021
, “
Exploiting Local Geometric Features in Vehicle Design Optimization With 3d Point Cloud Autoencoders
,” 2021 IEEE Congress on Evolutionary Computation (CEC), pp.
514
521
.
21.
Trinh
,
T. L.
,
Chen
,
F.
,
Nanri
,
T.
, and
Akasaka
,
K.
,
2024
, “
3d Super-Resolution Model for Vehicle Flow Field Enrichment
,” Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp.
5826
5835
.
22.
Deng
,
J.
,
Dong
,
W.
,
Socher
,
R.
,
Li
,
L.-J.
,
Li
,
K.
, and
Fei-Fei
,
L.
,
2009
, “
Imagenet: A Large-Scale Hierarchical Image Database
,” 2009 IEEE Conference on Computer Vision and Pattern Recognition,
IEEE
, pp.
248
255
.
23.
Chang
,
A. X.
,
Funkhouser
,
T.
,
Guibas
,
L.
,
Hanrahan
,
P.
,
Huang
,
Q.
,
Li
,
Z.
,
Savarese
,
S.
,
Savva
,
M.
,
Song
,
S.
,
Su
,
H.
, et al.,
2015
, “Shapenet: An Information-Rich 3d Model Repository,” preprint arXiv:1512.03012.
24.
Ashton
,
N.
,
Batten
,
P.
,
Cary
,
A.
, and
Holst
,
K.
,
2023
, “
Summary of the 4th High-lift Prediction Workshop Hybrid RANS/LES Technology Focus Group
,”
J. Aircraft
,
61
, pp.
1
30
.
25.
Ashton
,
N.
, and
van Noordt
,
W.
,
2022
, “
Overview and Summary of the First Automotive CFD Prediction Workshop: Drivaer Model
,”
SAE Int. J. Commercial Veh.
,
16
(
02-16-01-0005
).
26.
Jacob
,
S. J.
,
Mrosek
,
M.
,
Othmer
,
C.
, and
Köstler
,
H.
,
2022
, “
Deep Learning for Real-Time Aerodynamic Evaluations of Arbitrary Vehicle Shapes
,”
SAE Int. J. Passenger Veh. Syst.
,
15
(
2
), pp.
77
90
.
27.
Kontou
,
M. G.
,
Asouti
,
V. G.
, and
Giannakoglou
,
K. C.
,
2023
, “
DNN Surrogates for Turbulence Closure in CFD-Based Shape Optimization
,”
Appl. Soft Comput.
,
134
, p.
110013
.
28.
Heft
,
A. I.
,
Indinger
,
T.
, and
Adams
,
N. A.
,
2012
, “Introduction of a New Realistic Generic Car Model for Aerodynamic Investigations,” Technical Report, SAE Technical Paper.
29.
Deng
,
J.
,
Li
,
X.
,
Xiong
,
H.
,
Hu
,
X.
, and
Ma
,
J.
, “
Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence
,”
IJCAI '24
,
Jeju, South Korea
,
Aug. 3–9
.
30.
Wu
,
H.
,
Luo
,
H.
,
Wang
,
H.
,
Wang
,
J.
, and
Long
,
M.
,
2024
, “
Proceedings of the 41st International Conference on Machine Learning
,”
ICML
,
Vienna, Austria
,
July 21–27
.
31.
Menter
,
F. R.
,
Kuntz
,
M.
, and
Langtry
,
R.
,
2003
, “
Ten Years of Industrial Experience With the SST Turbulence Model
,”
Turbulence Heat Mass Transfer
,
4
(
1
), pp.
625
632
.
32.
Spalding
,
D. B.
,
1974
, “
The Numerical Computation of Turbulent Flow
,”
Comput. Methods Appl. Mech. Eng.
,
3
, p.
269
.
33.
Wilcox
,
D. C.
,
1998
ṠEP,
Turbulence Modeling for CFD
, Vol. 2,
DCW Industries La Canada
,
CA
.
34.
Greenshields
,
C.
,
2023
,
OpenFOAM V11 User Guide
,
The OpenFOAM Foundation
,
London
.
35.
Ahmed
,
S. R.
,
Ramm
,
G.
, and
Faltin
,
G.
,
1984
, “
Some Salient Features of the Time-Averaged Ground Vehicle Wake
,”
SAE Trans.
,
93
, pp.
473
503
.
36.
Pavia
,
G.
, and
Passmore
,
M.
,
2018
, “
Characterisation of Wake Bi-stability for a Square-Back Geometry With Rotating Wheels
,” Progress in Vehicle Aerodynamics and Thermal Management: 11th FKFS Conference, Stuttgart, Sept. 26–27, 2017,
Springer
, pp.
93
109
.
37.
Drivaer Model Geometry
,
2024
, https://www.epc.ed.tum.de/en/aer/research-groups/automotive/drivaer/geometry/, Accessed May 21, 2024.
38.
Cogotti
,
A.
,
1998
, “
A Parametric Study on the Ground Effect of a Simplified Car Model
,”
SAE Trans.
,
107
, pp.
180
204
.
39.
Damblin
,
G.
,
Couplet
,
M.
, and
Iooss
,
B.
,
2013
, “
Numerical Studies of Space-Filling Designs: Optimization of Latin Hypercube Samples and Subprojection Properties
,”
J. Simul.
,
7
(
4
), pp.
276
289
.
40.
Wieser
,
D.
,
Schmidt
,
H.-J.
,
Mueller
,
S.
,
Strangfeld
,
C.
,
Nayeri
,
C.
, and
Paschereit
,
C.
,
2014
, “
Experimental Comparison of the Aerodynamic Behavior of Fastback and Notchback Drivaer Models
,”
SAE Int. J. Passenger Cars-Mech. Syst.
,
7
(
2014-01-0613
), pp.
682
691
.
41.
Elrefaie
,
M.
,
Hüttig
,
S.
,
Gladkova
,
M.
,
Gericke
,
T.
,
Cremers
,
D.
, and
Breitsamter
,
C.
,
2024
, “
On-Site Aerodynamics Using Stereoscopic PIV and Deep Optical Flow Learning
,”
Exp. Fluids
,
65
(
12
), pp.
1
20
.
42.
Romor
,
F.
,
Tezzele
,
M.
,
Mrosek
,
M.
,
Othmer
,
C.
, and
Rozza
,
G.
,
2023
, “
Multi-fidelity Data Fusion Through Parameter Space Reduction With Applications to Automotive Engineering
,”
Int. J. Numer. Methods Eng.
,
124
(
23
), pp.
5293
5311
.
43.
Shen
,
Y.
,
Patel
,
H. C.
,
Xu
,
Z.
, and
Alonso
,
J. J.
,
2024
, “
Application of Multi-fidelity Transfer Learning With Autoencoders for Efficient Construction of Surrogate Models
,” AIAA SCITECH 2024 Forum, p.
0013
.
44.
Fan
,
H.
,
Su
,
H.
, and
Guibas
,
L. J.
,
2017
, “
A Point Set Generation Network for 3d Object Reconstruction From a Single Image
,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.
605
613
.
45.
Abbas
,
A.
,
Rafiee
,
A.
,
Haase
,
M.
, and
Malcolm
,
A.
,
2022
, “
Geometrical Deep Learning for Performance Prediction of High-speed Craft
,”
Ocean Eng.
,
258
, p.
111716
.
46.
Pfaff
,
T.
,
Fortunato
,
M.
,
Sanchez-Gonzalez
,
A.
, and
Battaglia
,
P. W.
,
2021
, “
International Conference on Learning Representations
,”
ICLR
,
Vienna, Austria
.
47.
Rios
,
T.
,
Sendhoff
,
B.
,
Menzel
,
S.
,
Back
,
T.
, and
Stein
,
B. V.
,
2019
, “On the Efficiency of a Point Cloud Autoencoder as a Geometric Representation for Shape Optimization,”
Institute of Electrical and Electronics Engineers Inc
., pp.
791
798
.
48.
Rios
,
T.
,
Stein
,
B. V.
,
Back
,
T.
,
Sendhoff
,
B.
, and
Menzel
,
S.
,
2021
, “Point2ffd: Learning Shape Representations of Simulation-Ready 3d Models for Engineering Design Optimization,”
Institute of Electrical and Electronics Engineers Inc
., pp.
1024
1033
.
49.
Rios
,
T.
,
Wollstadt
,
P.
,
Stein
,
B. V.
,
Back
,
T.
,
Xu
,
Z.
,
Sendhoff
,
B.
, and
Menzel
,
S.
,
2019
, “Scalability of Learning Tasks on 3d CAE Models Using Point Cloud Autoencoders,”
Institute of Electrical and Electronics Engineers Inc
., pp.
1367
1374
.
50.
Sanchez-Gonzalez
,
A.
,
Godwin
,
J.
,
Pfaff
,
T.
,
Ying
,
R.
,
Leskovec
,
J.
, and
Battaglia
,
P.
,
2020
, “
Learning to Simulate Complex Physics With Graph Networks
,” International Conference on Machine Learning,
PMLR
, pp.
8459
8468
.
51.
Qi
,
C. R.
,
Su
,
H.
,
Mo
,
K.
, and
Guibas
,
L. J.
,
2017
, “
Pointnet: Deep Learning on Point Sets for 3d Classification and Segmentation
,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.
652
660
.
52.
Kingma
,
D. P.
, and
Ba
,
J.
,
2014
, “Adam: A Method for Stochastic Optimization,” preprint arXiv:1412.6980.
53.
Paszke
,
A.
,
Gross
,
S.
,
Chintala
,
S.
,
Chanan
,
G.
,
Yang
,
E.
,
DeVito
,
Z.
,
Lin
,
Z.
,
Desmaison
,
A.
,
Antiga
,
L.
, and
Lerer
,
A.
,
2017
,
NIPS 2017 Workshop Autodiff
.
You do not currently have access to this content.