LQG Controller Design for Hover Flight Based on Model Helicopter System Identification, Using X-Plane

Please enable javascript in your browser.

Yousefi Bonab, Soheil | 2019

521 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 52002 (45)
University: Sharif University of Technology
Department: Aerospace Engineering
Advisor(s): Banazadeh, Afshin
Abstract:
The subject of this study is design of a linear quadratic Gaussian controller based on identification of dynamic behavior of unmanned model helicopter “Raptor” in its hover flight. To achieve this end, first, the helicopter linear transfer functions are identified with frequency response method by applying specific inputs to the model of Raptor in the X-Plane software and receiving system outputs via Simulink. Next, using the identified dynamic model of the helicopter, an optimum linear controller is designed for hover flight. As an optimum linear controller needs all the state variables, a Kalman filter is utilized to estimate them, using outputs of various sonsors such as GPS and gyro. Finally, the optimum LQR controller is applied to the nonlinear model of the rotorcraft in the X-Plane software
Keywords:
System Identification ; Model Helicopter Flight Simulation ; Controller Design ; Linear Quadratic Gaussian (LQG)Controller ; Hover Flight

No TOC