EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 2.5

Aerodynamic Parameter Estimation using Adaptive Forgetting Factor Recurrent Neural Network with Lyapunov Stability Analysis

Sara Mohan George, J. R. Raol

Defence Science Journal · 22.06.2026 · 76 (4) · 477-487

DOI: 10.14429/dsj.21510 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Defence Science Journal, online erschienen am 22.06.2026. Verfasst von Sara Mohan George und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 2.5).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate estimation of aircraft aerodynamic and stability parameters plays a very important role in flight dynamics modeling, control system design and performance evaluation, especially in defence aviation and unmanned aerial vehicles (UAVs). Parameters such as lift, drag, moment coefficients, and control derivatives are typically derived from flight tests, but conducting such experiments can be expensive, risky, or restricted in defence settings. Existing methods, including gradient-based recurrent neural networks (RNN-G), face limitations in handling fast-changing, nonlinear dynamics owing to their static memory structures. This study proposes an Adaptive Forgetting Factor Recurrent Neural Network (AFF-RNN) that dynamically adjusts the forgetting factor in response to evolving error trends. The method was implemented in MATLAB and tested on simulated flight scenarios to assess its performance under different dynamic conditions. Results show that AFF-RNN achieved up to 25 % improvement in estimation accuracy compared to RNN-G. A stability analysis based on a normalized Lyapunov energy functional confirmed asymptotic convergence, ensuring reliability for safety-critical applications. AFF-RNN demonstrates consistent performance in aircraft parameter estimation with its flexibility, robustness, and theoretical assurances. It can also be used as a basis for future extensions to actual flight data in defense applications.

Zitieren

APA

George, S. M., & Raol, J. R. (2026). Aerodynamic Parameter Estimation using Adaptive Forgetting Factor Recurrent Neural Network with Lyapunov Stability Analysis. Defence Science Journal, 76(4), 477-487. https://doi.org/10.14429/dsj.21510

Vancouver

George SM, Raol JR. Aerodynamic Parameter Estimation using Adaptive Forgetting Factor Recurrent Neural Network with Lyapunov Stability Analysis. Defence Science Journal. 2026;76(4):477-487. doi:10.14429/dsj.21510

BibTeX
@article{George2026,
  author = {George, Sara Mohan and Raol, J. R.},
  title = {Aerodynamic Parameter Estimation using Adaptive Forgetting Factor Recurrent Neural Network with Lyapunov Stability Analysis},
  journal = {Defence Science Journal},
  year = {2026},
  volume = {76},
  number = {4},
  pages = {477--487},
  doi = {10.14429/dsj.21510}
}
Kostenlos starten →