Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/79539
Title: Wind-resistant hover control and wind field estimation of a VTOL tail-sitter UAV
Authors: Sun, Jingxuan
Advisors: Wen, Chih-yung (ME)
Keywords: Drone aircraft -- Control systems
Drone aircraft -- Automatic control
Issue Date: 2018
Publisher: The Hong Kong Polytechnic University
Abstract: Tail-sitter unmanned aerial vehicles (UAVs) have been studied and discussed by many researchers because of their simple structure and excellent mobility. However, it still needs improvement before it can be used in practical applications. A dual-rotor tail-sitter is studied in this thesis. A Hardware-in-loop (HIL) simulation environment was developed for the dual-rotor tail-sitter. The 6 degrees-of-freedom (DOF) equations of motion (EOM) were driven to model the vehicle. Component breakdown method was adopted to solve the partial flow condition problem. The aerodynamic characteristics were obtained from the wind tunnel experiments. The dynamic simulator with the capability of real-time simulation was thus built. The HIL simulation environment supports the open-source flight controller and Robot Operating System (ROS). The flight control system communicated with the simulator with the User-defined protocol (UDP). The HIL simulation environment was evaluated with both indoor and outdoor flight tests and demonstrated its reliability. A control strategy was proposed and developed to improve the flight performance of the tail-sitter hovering in the prevailing wind conditions. In this strategy, the tail-sitter is controlled to reduce the immersed area in the wind. A wind vane and an angular sensor were installed on the vehicle to measure the direction of the coming flow and communicated to the flight controller. The finite impulse response (FIR) filter is applied to smooth the measured data. The developed system has been validated in flight tests, which include a change of 90 degrees in wind direction. The wind direction was accurately measured by the onboard system, and the vehicle successfully captured the motion. The flight data was compared with the original controller and the advantages of the proposed method are shown. The method of wind field estimation method for tail-sitter in both hover and cruise phase is presented in this thesis. Considering the inaccuracy of low-cost sensors and limited loading capability of the small UAV, a Global Positioning System (GPS) sensor, a pitot tube and an Inertial measurement unit (IMU) are needed, which are commonly installed on small tail-sitter UAVs. The combination of the wind triangle and aerodynamic relationship was adopted in this study for the wind estimation. Both the extended Kalman filter (EKF) and the Unscented Kalman filter (UKF) were adopted in the proposed method and compared. The proposed method caught the prevailing wind in a few time steps and shows accurate results in both hover and cruise phase.
Description: xiv, 112 pages : color illustrations
PolyU Library Call No.: [THS] LG51 .H577P ME 2018 Sun
URI: http://hdl.handle.net/10397/79539
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
991022164555603411_link.htmFor PolyU Users167 BHTMLView/Open
991022164555603411_pira.pdfFor All Users (Non-printable)10.21 MBAdobe PDFView/Open
Show full item record
PIRA download icon_1.1View/Download Contents

Page view(s)

11
Citations as of Dec 10, 2018

Download(s)

2
Citations as of Dec 10, 2018

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.