Recurrent Neural Network for Estimation of Aerodynamic Parameters

In this paper, Recurrent Neural Network (RNN) is proposed for the estimation of aerodynamic derivatives of an aircraft. The simulated flight data for short period mode of research aircraft, Hansa is used to train the RNN. The trained RNN is then used to predict the output for a given input file. The...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Conference on Signal Processing and Integrated Networks (Online) s. 150 - 155
Hlavní autoři: Mahajan, Hardik K., Kaur, Jaspreet, Singh, S., Banerjee, S.
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 26.08.2021
Témata:
Aerodynamics
Backpropagation
Delta Method
Estimation
Feed Forward Back Propagation Neural Network
Force
Parameter Estimation
Recurrent Neural Network
Recurrent neural networks
Signal processing
Signal processing algorithms
ISSN:2688-769X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In this paper, Recurrent Neural Network (RNN) is proposed for the estimation of aerodynamic derivatives of an aircraft. The simulated flight data for short period mode of research aircraft, Hansa is used to train the RNN. The trained RNN is then used to predict the output for a given input file. The predicted data is then used for the estimation of aerodynamic derivatives using the Delta Method. The results obtained using RNN are compared to the results obtained using Feed Forward Back propagation algorithm (FFBP). It is found that the derivatives obtained using RNN are very close to true values of derivatives with lesser standard deviation as compared to derivatives obtained using FFBP algorithm. The RNN is also validated by adding the various percentages of noises in the simulated flight data. The results increase level of confidence and suggest that the RNN can be used advantageously to estimate aerodynamic derivatives of an aircraft from real flight data.
ISSN:2688-769X
DOI:10.1109/SPIN52536.2021.9566075