Presentation Title

Flight Testing, Data Collection, and System Identification of a Multicopter UAV

Start Date

November 2016

End Date

November 2016

Location

Watkins 1111

Type of Presentation

Oral Talk

Abstract

This presentation talks about the flight testing, data collection, data processing, and system identification of a multicopter UAV (Unmanned Aerial Vehicle). A flight dynamics model of the multicopter is developed using a system identification technique. The collected flight data is converted to a frequency response using CIFER (Comprehensive Identification from FrEquency Response) software. The frequency response is then used in the identification of transfer function and state-space mathematical models of the multicopter in hovering flight. The collected data during flight tests was first processed using MATLAB and Simulink software. The processed data was then used for the model identification. Response of the closed-loop transfer functions matched the flight data very well while the response of the transfer functions obtained using open-loop data, through similar analysis, is inconsistent due to noise and disturbances. Work is underway to remove noise from the data using noise filters and creating a state-space models. Also, more flight test will be conducted to collect data for forward flights so that that mathematical models of the multicopter can be developed for forward flight. Simulation results compared with flight data will be presented.

This document is currently not available here.

Share

COinS
 
Nov 12th, 11:45 AM Nov 12th, 12:00 PM

Flight Testing, Data Collection, and System Identification of a Multicopter UAV

Watkins 1111

This presentation talks about the flight testing, data collection, data processing, and system identification of a multicopter UAV (Unmanned Aerial Vehicle). A flight dynamics model of the multicopter is developed using a system identification technique. The collected flight data is converted to a frequency response using CIFER (Comprehensive Identification from FrEquency Response) software. The frequency response is then used in the identification of transfer function and state-space mathematical models of the multicopter in hovering flight. The collected data during flight tests was first processed using MATLAB and Simulink software. The processed data was then used for the model identification. Response of the closed-loop transfer functions matched the flight data very well while the response of the transfer functions obtained using open-loop data, through similar analysis, is inconsistent due to noise and disturbances. Work is underway to remove noise from the data using noise filters and creating a state-space models. Also, more flight test will be conducted to collect data for forward flights so that that mathematical models of the multicopter can be developed for forward flight. Simulation results compared with flight data will be presented.