A physics-inspired-data-driven model of eVTOL in-ground-effect thrust based on meta-learning and BEMT

Abstract

Electric vertical take-off and landing (eVTOL) aircraft with ducted fans have high efficiency, high reliability, and low noise, and have great potential to become the majority of future urban air mobility. Nonetheless, the ground effect will cause fluctuations in ducted fan thrust during take-off and landing conditions. Conventional physical models fail to reflect its transient and non-linear nature. In this paper, a physics-inspired-data-driven (PIDD) model for predicting the in-ground-effect (IGE) thrust of ducted fans is presented. The total thrust is treated as the composition of average thrust and transient thrust. For the average part, hypotheses of ideal twist and exponential function are adopted based on the structure of blade element momentum theory (BEMT). For the transient part, a classification task based on cross entropy is added to the meta-learning algorithm. Altitude related and altitude unrelated items are constructed to obtain the PIDD model and enhance its generalization ability. The features of the model are rotary speed, voltage, and current, while the label is IGE thrust. Few-shot training indicates that the proposed model can predict transient thrust accurately. With mean error on training and testing sets is 1.1% and 0.7% respectively, the PIDD model outperforms the conventional physical model by 4.6% and 1.7% and the data-driven model by 1.5% and 2.7%. The PIDD model constructed succeeds in predicting transient IGE thrust and provides new ideas for relevant research.

Keywords

Meta-learning / BEMT / eVTOL / Ground effect / Ducted fan

References

[1] Zhang Y, Qian Y. et al.. Progress and key technologies of flying cars. Journal of Automotive Safety and Energy, 2020, 11(01): 1-16(in Chinese)

[2] Luo Y, Ai T, He Y. et al.. Numerical investigation on unsteady characteristics of ducted fans in ground effect. Chin J Aeronaut, 2023, 36: 79-95.

[3] Ai T, Xu B, Xiang C. et al.. Aerodynamic analysis and control for a novel coaxial ducted fan aerial robot in ground effect. Int J Adv Rob Syst, 2020, 17: 1729881420953026.

[4] Ai T, Fan W, Xu B. et al.. Aerodynamic analysis and modeling of coaxial ducted fan aircraft with the ceiling effect. Engineering Applications of Computational Fluid Mechanics, 2021, 15: 1563-1584.

[5] Luo Y, Ai T, He Y. et al.. Aerodynamic analysis on unsteady characteristics of a ducted fan hovering in ceiling effect. Engineering Applications of Computational Fluid Mechanics, 2023, 17: 2196327.

[6] Vong CH, Ryan K, Chung H. Integral backstepping position control for quadrotors in tunnel-like confined environments. International Conference on Robotics and Automation (ICRA), 2019, 2019: 6425-6431.

[7] Matus-Vargas A, Rodriguez-Gomez G, Martinez-Carranza J. Ground effect on rotorcraft unmanned aerial vehicles: a review. Intel Serv Robot, 2021, 14: 99-118.

[8] O’Connell M, Shi G, Shi X, et al. (2022) Neural-Fly enables rapid learning for agile flight in strong winds. Sci Robotics, 2022, 7(66): eabm6597. https://doi.org/10.1126/scirobotics.abm6597

[9] Luo Y, He Y, Qian Y. et al.. Aerodynamic modeling of ducted fans under environmental disturbances. J Phys: Conf Ser, 2024, 2024012075

[10] Luo Y, Ai T, He Y et al (2024) Numerical analysis of wind effects on aerodynamic characteristics of a ducted fan. Chin J Aeronaut 37(5):263–280. https://doi.org/10.1016/j.cja.2024.02.002

[11] Qian Y, Luo Y, Hu X, Zeng Z, Zhang Y. Improving the performance of ducted fans for VTOL applications: A review. SCIENCE CHINA Technol Sci, 2022, 65(11): 2521-2541.

[12] Mahony R, Hamel T. Adaptive Compensation of Aerodynamic Effects during Takeoff and Landing Manoeuvres for a Scale Model Autonomous Helicopter. Eur J Control, 2001, 7(1): 43-57.

[13] Guenard N, Hamel T, Eck L (2006) Control laws for the tele operation Of an unmanned aerial vehicle known as an X4-flyer. In: 2006 IEEE/RSJ international conference on intelligent robots and systems, IEEE, Beijng

[14] Bartholomew J, Calway A, Mayol-Cuevas W (2015) Improving MAV control by predicting aerodynamic effects of obstacles. In: 2015 IEEE/RSJ international conference on intelligent robots and systems (IROS), IEEE, Hamburg

[15] Shi G, Shi X, O’Connell M et al (2019) Neural lander: stable drone landing control using learned dynamics. In: 2019 international conference on robotics and automation (ICRA), IEEE, Montreal

[16] Gao S, Franco CD, Carter D et al (2019) Exploiting ground and ceiling effects on autonomous UAV motion planning. In: 2019 international conference on unmanned aircraft systems (ICUAS), IEEE, Atlanta

[17] Liu M, Qian Y, Luo Y, Hao H, Liu Z, Zhao F. et al.. Lifecycle greenhouse gas emissions and energy cost analysis of flying cars with three different propulsion systems. J Clean Prod, 2022, 331. 129985

[18] Luo Y, Qian Y, Zeng Z et al (2021) Simulation and analysis of operating characteristics of power battery for flying car utilization. eTransportation. 8:100111. https://doi.org/10.1016/j.etran.2021.100111

[19] Deng S, Wang S, Zhang Z. Aerodynamic performance assessment of a ducted fan UAV for VTOL applications. Aerosp Sci Technol, 2020, 103. 105895

[20] Mi B-g. Numerical investigation on aerodynamic performance of a ducted fan under interferences from the ground, static water and dynamic waves. Aerosp Sci Technol, 2020, 100. 105821

[21] Garofano-Soldado A, Sanchez-Cuevas PJ, Heredia G, Ollero A. Numerical-experimental evaluation and modelling of aerodynamic ground effect for small-scale tilted propellers at low Reynolds numbers. Aerosp Sci Technol, 2022, 126. 107625

[22] Hsiao YH, Chirarattananon P (2018) Ceiling effects for surface locomotion of small rotorcraft. In: 2018 IEEE/RSJ international conference on intelligent robots and systems (IROS), IEEE, Madrid

[23] Pasquali C, Serafini J, Bernardini G, Milluzzo J, Gennaretti M. Numerical-experimental correlation of hovering rotor aerodynamics in ground effect. Aerosp Sci Technol, 2020, 106. 106079

[24] Pasquali C, Gennaretti M, Bernardini G, Serafini J. State-space dynamic inflow modelling for hovering rotors in fixed- and moving-ground effect. Aerosp Sci Technol, 2023, 140. 108414

[25] Jin Y, Fu Y, Qian Y, Zhang Y. A Moore-Greitzer Model for Ducted Fans in Ground Effect. Journal Of Applied Fluid Mechanics, 2020, 13(2): 693-701.

[26] He X, Leang KK (2019) A new quasi-steady in-ground effect model for rotorcraft unmanned aerial vehicles. No. ASME DSCC2019-9025

[27] Ryan T, Kim HJ (2012) Modelling of quadrotor ground effect forces via simple visual feedback and support vector regression. No. AIAA 2012-4833

[28] Salzmann T, Kaufmann E, Arrizabalaga J, Pavone M, Scaramuzza D, Ryll M. Real-Time Neural MPC: Deep Learning Model Predictive Control for Quadrotors and Agile Robotic Platforms. IEEE Robotics and Automation Letters, 2023, 8(4): 2397-2404.

[29] Mohajerin N, Waslander SL (2015) Modelling a quadrotor vehicle using a modular deep recurrent neural network. In: 2015 IEEE international conference on systems, IEEE, Hong Kong

[30] Finn C, Abbeel P, Levine S (2017) Model-agnostic meta-learning for fast adaptation of deep networks. In: 34th international conference on machine learning, Sydney

[31] Belkhale S, Li R, Kahn G et al (2021) Model-based meta-reinforcement learning for flight with suspended payloads. IEEE Robot Auto Letters 6(2):1471–1478. https://doi.org/10.1109/LRA.2021.3057046

RIGHTS & PERMISSIONS

The Author(s)