Long Term Learning Adaptive Neural Network Estimator Based Limit Detection

Date

2010-08-26

Author

Yavrucuk, İlkay

Metadata

Show full item record

Item Usage Stats

167
views

0
downloads

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 simulations. It is observed that by using recorded data in a stack can cancel out new modeling errors in a short time and results with better predictions of approaching limits compared to using instantaneous data only

Subject Keywords

Limit avoidance, Limit detection, Pilot cueing

URI

https://hdl.handle.net/11511/83455

DOI

https://doi.org/10.3182/20100826-3-tr-4015.00052

Conference Name

10th IFAC International Workshop on Adaptation and Learning in Control and Signal Processing (26 Ağustos - 28 Ekim 2010)

Collections

Department of Aerospace Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Concurrent Learning Enabled Adaptive Limit Detection for Active Pilot Cueing Yavrucuk, İlkay (2010-08-02) 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...
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...
Navigation algorithms and autopilot application for an unmanned airvehicle Kahraman, Eren; Alemdaroğlu, Hüseyin Nafiz; Nalbantoğlu, Volkan; Department of Aerospace Engineering (2010) This study describes the design and implementation of the altitude and heading autopilot algorithms for a fixed wing unmanned air vehicle and navigation algorithm for attitude and heading reference outputs. Algorithm development is based on the nonlinear mathematical model of Middle East Technical University Tactical Unmanned Air Vehicle (METU TUAV), which is linearized at a selected trim condition. A comparison of nonlinear and linear mathematical models is also done. Based on the linear mathematical model...
Numerical and experimantal analysis of flapping motion in hover, application to micro air vehicles Kurtuluş, Dilek Funda; Alemdaroğlu, Hüseyin Nafiz; Department of Aerospace Engineering (2005) The aerodynamics phenomena of flapping motion in hover are considered in view of the future Micro Air Vehicle applications. The aim of this work is to characterize the vortex dynamics generated by the wing in motion using direct numerical simulation and experimental analysis then to propose a simplified analytical model for prediction of the forces in order to optimize the parameters of the motion leading to maximum force. A great number of cases are investigated corresponding to different angles of attack,...
Small-size unmanned model helicopter guidance and control Karasu, Çağlar; Leblebicioğlu, Mehmet Kemal; Department of Electrical and Electronics Engineering (2004) The deployment of unmanned aerial vehicles (UAV) in military applications increased the research about them and the importance of them. The unmanned helicopters are the most agile and maneuverable vehicles among the unmanned aerial vehicles (UAV). The ability of hovering and low speed cruise makes them even more attractive. Such abilities supply more areas to deploy the usage of the unmanned helicopters like search & rescue, mapping, surveillance. Autonomy is the key property for these vehicles. In order to...

Citation Formats

İ. Yavrucuk, “Long Term Learning Adaptive Neural Network Estimator Based Limit Detection,” presented at the 10th IFAC International Workshop on Adaptation and Learning in Control and Signal Processing (26 Ağustos - 28 Ekim 2010), Antalya, Turkey, 2010, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/83455.