Aerodynamic Optimization of UAV Wings Using Machine Learning

Year : 2026 | Volume : 02 | Issue : 01 | Page : 1 7
    By

    Nidhi Sahu*,

  • Rohit Singh Rathor,

  • Nara Siman,

  1. Research Scholar, Department of Mechanical Engineering, Lingayas Vidyapeeth Nauchali, Faridabad, Haryana, India
  2. Head of Department, Department of Mechanical Engineering, N C University, Faridabad, Haryana, India
  3. Head of Department, Department of Mechanical Engineering, N C University, Faridabad, Haryana, India

Abstract

Unmanned Aerial Vehicles (UAVs) are increasingly deployed across defense, transportation, agriculture, and environmental monitoring, demanding improved aerodynamic efficiency to enhance endurance, stability, and payload capacity. Traditional aerodynamic optimization approaches, relying on computational fluid dynamics (CFD) simulations and wind tunnel experiments, are often time-consuming and computationally expensive. This study proposes a machine learning (ML)-driven framework for the aerodynamic optimization of UAV wing geometries, aiming to significantly reduce design cycles while improving aerodynamic performance parameters such as lift-to-drag ratio, stall characteristics, and pressure distribution. A comprehensive dataset is generated through parametric CFD simulations involving variations in airfoil shape, angle of attack, camber, thickness, and aspect ratio. Feature engineering techniques are applied to extract dominant aerodynamic influences, and multiple ML algorithms—Random Forest Regression, Gradient Boosting, Artificial Neural Networks, and Support Vector Machines—are trained to predict aerodynamic coefficients with high accuracy. The best-performing model is integrated with a multi-objective optimization algorithm to automatically identify optimal wing configurations. Validation is performed through additional CFD runs and selected wind tunnel tests to ensure reliability. Results demonstrate that the ML-based approach reduces computational effort by nearly 60–70% compared to conventional CFD-only optimization, while achieving the significant improvement in lift-to-drag ratio and flow uniformity. The proposed methodology highlights the potential of data-driven design tools for accelerating UAV development, enabling engineers to rapidly explore large design spaces with enhanced precision. This research contributes to the growing integration of artificial intelligence in aerodynamic design and provides a scalable framework for future autonomous aerial platforms.

Keywords: Aerodynamic optimization, airfoil shape optimization, computational fluid dynamics (CFD), machine learning (ML), unmanned aerial vehicles (UAVs), wing geometry design

[This article belongs to International Journal on Drones ]

How to cite this article:
Nidhi Sahu*, Rohit Singh Rathor, Nara Siman. Aerodynamic Optimization of UAV Wings Using Machine Learning. International Journal on Drones. 2026; 02(01):1-7.
How to cite this URL:
Nidhi Sahu*, Rohit Singh Rathor, Nara Siman. Aerodynamic Optimization of UAV Wings Using Machine Learning. International Journal on Drones. 2026; 02(01):1-7. Available from: https://journals.stmjournals.com/ijd/article=2026/view=239854


References

 

  1. Karali H, Demirezen MU, Yukselen MA, Inalhan G. Design of a deep learning based nonlinear aerodynamic surrogate model for UAVs. In: AIAA, editor. AIAA Scitech 2020 Forum. 1st edition. Reston, USA: AIAA; 2020. pp. 1288.
  2. Karali H, Inalhan G, Tsourdos A. Advanced UAV Design Optimization Through Deep Learning-Based Surrogate Models. In: MDPI, editor. Aerospace. 11th edition. Basel, Switzerland: MDPI; 2024. pp. 669.
  3. Teimourian A, Rohacs D, Dimililer K, Teimourian H, Yildiz M, Kale U. Airfoil aerodynamic performance prediction using machine learning and surrogate modeling. In: Elsevier, editor. Heliyon. 10th edition. Amsterdam, Netherlands: Elsevier; 2024. pp. e29145.
  4. Minaev E, Quijada Pioquinto JG, Shakhov V, Kurkin E, Lukyanov O. Airfoil optimization using deep learning models and evolutionary algorithms for the case large-endurance UAVs design. In: MDPI, editor. Drones. 8th edition. Basel, Switzerland: MDPI; 2024. pp. 570.
  5. Wang L, Zhang H, Wang C, Tao J, Lan X, Sun G, Feng J. A review of intelligent airfoil aerodynamic optimization methods based on data-driven advanced models. In: MDPI, editor. Mathematics. 12th edition. Basel, Switzerland: MDPI; 2024. pp. 1417.
  6. Ebrahimi MJ, Pasandidehfard M, Esmaeili A, Moghimi Esfandabadi MH. Optimization of aerodynamic design of a winged UAV through genetic algorithms and large eddy simulation. In: Hindawi, editor. International Journal of Aerospace Engineering. 1st edition. London, UK: Hindawi; 2024. pp. 2698950.
  7. Kieszek R, Majcher M, Syta B, Kozakiewicz A. Feed-forward artificial neural network as surrogate model to predict lift and drag coefficient of NACA airfoil and searching of maximum lift-to-drag ratio. In: PTMTS, editor. Journal of Theoretical and Applied Mechanics. 62nd edition. Warsaw, Poland: PTMTS; 2024. pp. 155-168.
  8. Li J, Cai J. Massively multipoint aerodynamic shape design via surrogate-assisted gradient-based optimization. In: AIAA, editor. AIAA Journal. 58th edition. Reston, USA: AIAA; 2020. pp. 1949–1963.
  9.  Zhang Z, Zhao P, Zhang Z, Sun Q, Zhang J, Deng Z. Machine learning-enhanced optimization of rotor blades for rotary-wing Mars UAVs through coupled CFD simulation. In: Elsevier, editor. Aerospace Science and Technology. 158th edition. Amsterdam, Netherlands: Elsevier; 2025. pp. 109858.
  10.  Pliakos C, Efrem G, Terzis D, Panagiotou P. An Automated Framework for Streamlined CFD-Based Design and Optimization of Fixed-Wing UAV Wings. In: MDPI, editor. Algorithms. 18th edition. Basel, Switzerland: MDPI; 2025. pp. 186.
  11. Song H, Sheng Z, Zhang B, Deng Y, Ji W, Yu J. Machine learning control of airfoil lift increase using co-flow jet. In: AIP, editor. Physics of Fluids. 37th edition. Melville, USA: AIP Publishing; 2025. pp. 11.
  12.  Hammouda NG, Khan R, Mostafa L, Musa VA, Rajab H, Singh NS, Aich W, Hajlaoui K. Application of deep reinforcement learning for aerodynamic control around an angled airfoil via synthetic jet. In: Nature, editor. Scientific Reports. 15th edition. London, UK: Nature Publishing Group; 2025. pp. 1-15.
  1. Teimourian A, Rohacs D, Dimililer K, Teimourian H, Yildiz M, Kale U. Airfoil aerodynamic performance prediction using machine learning and surrogate modeling. In: Elsevier, editor. Heliyon. 10th edition. Amsterdam, Netherlands: Elsevier; 2024. pp. e29145.

 

 

 

Regular Issue Subscription Review Article
Volume 02
Issue 01
Received 27/10/2025
Accepted 03/02/2026
Published 13/02/2026
Publication Time 109 Days
6

Login


My IP

PlumX Metrics