Tracking Control of Quadrotor with Computer Vision Implemented in ROS

碩士 === 國立宜蘭大學 === 機械與機電工程學系碩士班 === 105 === The purpose of this study is to design a control system for a quadrotor using Robot Operating System(ROS). The control system is able to tune the quadrotor‘s attitude automatically and to make the quadrotor hovering over a target point. In this study, a qua...

Full description

Bibliographic Details
Main Authors: Yi-Lun Hsieh, 謝宜倫
Other Authors: Bih-Sheng Hsu
Format: Others
Language:zh-TW
Published: 2017
Online Access:http://ndltd.ncl.edu.tw/handle/rpsr7z
Description
Summary:碩士 === 國立宜蘭大學 === 機械與機電工程學系碩士班 === 105 === The purpose of this study is to design a control system for a quadrotor using Robot Operating System(ROS). The control system is able to tune the quadrotor‘s attitude automatically and to make the quadrotor hovering over a target point. In this study, a quadrotor with a flight controller MultiWii, a single board computer(SBC) Raspberry Pi, and Pi Camera is used. The SBC processes images captured from the camera and recognizes the target position. The control commands are passed to the quadrotor via ROS from Simulink, which is also utilized to analyse data and to tune the parameters of the control system. The controller is designed with a fuzzy PID using the Fuzzy Logic Toolbox of Matlab/Simulink, to verify the possibility and effectiveness of building a visual tracking control system with ROS.