One of the main research focus areas of the WVU Flight Control Systems Laboratory (FCSL) is the increase of flight safety through the implementation of fault tolerant control laws. For some fault tolerant flight control approaches with adaptive control laws, the availability of accurate post failure aircraft models improves performance. While look-up tables of aircraft models can be created for failure conditions, they may fail to account for all possible failure scenarios. Thus, a real-time parameter identification program eliminates the need to have predefined models for all potential failure scenarios. The goal of this research was to identify the dimensional stability and control derivatives of the WVU Phastball UAV in flight using a frequency domain based real-time parameter identification (PID) approach.
The data necessary for this project was gathered using the WVU Phastball UAV, a radio-controlled aircraft designed and built by the FCSL for fault tolerant control research. Maneuvers designed to excite the natural dynamics of the aircraft were implemented by the pilot or onboard computer during the steady state portions of flights. The data from these maneuvers was used for this project.
The project was divided into three main parts: 1) off-line time domain PID, 2) off-line frequency domain PID, and 3) an onboard frequency domain PID. The off-line parameter estimation programs, in both frequency domain and time domain, utilized the well known Maximum Likelihood Estimator with Newton-Raphson minimization with starting values estimated from a Least-Squares Estimate of the non-dimensional stability and control derivatives. For the frequency domain approach, both the states and inputs were first converted to the frequency domain using a Fourier integral over the frequency range in which the rigid body aircraft dynamics are found. The final phase of the project was a real-time parameter estimation program to estimate the dimensional stability and control derivatives onboard the Phastball aircraft. A frequency domain formulation of the least-squares estimation process was used because of its low computational and memory requirements and robustness to measurement noise and sensor information dropouts. Most of the onboard parameter estimates obtained converge to the values determined using the off-line parameter estimation programs (though a few typically show a bias) within four to six seconds for longitudinal estimates and four to eight seconds for the later estimates. For the experiments conducted, the real-time parameter estimates did not diverge after the conclusion of the maneuver.
|Commitee:||Gururajan, Srikanth, Perhinschi, Mario, Seanor, Brad|
|School:||West Virginia University|
|Department:||Mechanical and Aerospace Engineering|
|School Location:||United States -- West Virginia|
|Source:||MAI 51/05M(E), Masters Abstracts International|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be