Concurrent Learning Enabled Adaptive Limit Detection for Active Pilot Cueing

Date

2010-08-02

Author

Yavrucuk, İlkay

Metadata

Show full item record

Item Usage Stats

154
views

0
downloads

Adaptive dynamic models are commonly used to estimate allowable control travel and the proximity to a limiting flight condition in the design of advanced envelope protection algorithms for fly-by-wire aircraft. In this paper linear models are compensated with adaptive neural networks to build adaptive models of relevant aircraft dynamics. Neural networks are trained online not only using instantaneous sensor data, but also a stack of data collected during flight. The additional stack is made up of transient data as well as a collection of data representing steady state conditions. It is observed that using time history data stacks for online neural network training provides more accurate dynamic trim and control limit predictions compared to using instantaneous data only. Simulation results for a fixed wing aircraft during maneuvers, show comparisons between the different adaptation schemes.

URI

https://hdl.handle.net/11511/73834

DOI

https://doi.org/10.2514/6.2010-7945

Conference Name

AIAA Guidance, Navigation and Control Conference

Collections

Department of Aerospace Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Concurrent Learning Enabled Adaptive Limit Detection for Active Pilot Cueing Gursoy, Gonenc; Yavrucuk, İlkay (2014-09-01) Neural-network-based adaptive dynamic models are commonly used to estimate allowable control travel and the proximity to a limiting flight condition in the design of advanced envelope protection algorithms for fly-by-wire aircraft. In this paper, linear models are compensated with adaptive neural networks, which use instantaneous sensor data as well as past flight history information for concurrent learning. A law for collecting appropriate training data into the history stack is established. It is observed...
Long Term Learning Adaptive Neural Network Estimator Based Limit Detection Yavrucuk, İlkay (null; 2010-08-26) Dynamic adaptive models are commonly used to estimate allowable control travel and the proximity to a limiting flight condition in the design of advanced envelope protection algorithms for fly by wire aircraft. In this paper linear models are compensated with adaptive neural networks to build adaptive models of relevant aircraft dynamics. A stack of data collected during flight is used to update the network weights online. The data stack is made up of instantaneously measured data and recorded data during s...
Online Dynamic Trim and Control Limit Estimation Yavrucuk, İlkay (American Institute of Aeronautics and Astronautics (AIAA), 2012-9) The online estimation of a maneuvering steady-state condition of an aircraft, called the dynamic trim, is used to estimate the allowable control travel during flight, a key information in pilot cueing for envelope limit protection. In this paper a new methodology is presented where adaptive models are used to estimate online local dynamic trim conditions, while requiring very limited a priori vehicle information. Adaptive neural networks are employed to enable online learning. The models are used to estimat...
Direct Adaptive Limit and Control Margin Estimation with Concurrent Learning Gursoy, Gonenc; Yavrucuk, İlkay (American Institute of Aeronautics and Astronautics (AIAA), 2016-6) In this paper, two vital signals to enable flight envelope protection, namely the onset to the flight envelope (limit margin) and the available control travel to reach the limit boundary (control margin), are estimated using improved noniterative adaptive neural-network-based approximate models. The adaptive elements use current and past information (concurrent learning) and have guaranteed signal bounds. Minimum singular value maximization is used to record data for concurrent learning. Results showed bett...
Multi-fidelity aerodynamic dataset generation of a fighter aircraft Kurt, Huseyin Burak; Millidere, Murat; Gomec, Fazıl Selcuk; Uğur, Ömür (2021-01-01) High-fidelity aerodynamic dataset generation is one of the most significant components of the aircraft flight simulation, and, it is a time consuming and costly process. Data fusion techniques suggest that, instead of using high fidelity data for entire aerodynamic dataset, an incorporating combination of high-fidelity and low-fidelity data is a more cost-effective one. The objective of data fusion is to obtain high-fidelity dataset accuracy by combining less amount of high-fidelity dataset and more amount ...

Citation Formats

İ. Yavrucuk, “Concurrent Learning Enabled Adaptive Limit Detection for Active Pilot Cueing,” presented at the AIAA Guidance, Navigation and Control Conference, 2010, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/73834.