Indoor Localization and Flying Control of a Quadrotor Based on Kalman Filter

碩士 === 國立臺北科技大學 === 電機工程系 === 106 === In recent years, the research and development of unmanned aerial vehicle has been a popular research topic in the industry and academia, the types of research is very extensive. The chief reason for Attract the researchers to put into the realm with a developmen...

Full description

Bibliographic Details
Main Authors: Yuan-Chi Lin, 林源麒
Other Authors: 黃正民
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/xk9gb4
Description
Summary:碩士 === 國立臺北科技大學 === 電機工程系 === 106 === In recent years, the research and development of unmanned aerial vehicle has been a popular research topic in the industry and academia, the types of research is very extensive. The chief reason for Attract the researchers to put into the realm with a development potential in succession is the weight of quadcopter is lighter, the physical volume is small and dynamic sensitivity is high. This thesis controls with quadrotor roles, with the indoor localization system of ultra-wideband to get a coordinates of the quadrotor. This thesis designs the program to correct the Roll- axis and the Pitch- axis on the controller of the quadrotor. And developed a set of platform which completed the PID’s controller to control experiment of the quadrotor for quadrotor. According to different angles, recorded and analyzed experiment of phenomenon in more detail or attitude control. Even when the quadrotor stops to hover and it is subjected to the outside interference, it can return to the stable flight and reach the stable system status in a short time. Localization system of Ultra-wideband has shortest impulse time, it suits much to draw on accurate Localization indoor. It make use of the Time of Arrival (TOA) to position to get the position of quadrotor. When It combines with extended Kalman filter and probabilistic data association Filter, it will raise effectively the positioning precision .After multiple experiments and analytical results in the indoor scene of different heights. The results of experiments have corroborated to proposed that indoor localization system could estimate the position of the quadrotor correctly. Besides, while the quadrotor is flying indoors, collecting information from different places through a variety of sensor on embedded system platform rests on itself. At any moment we understand that information variation with time in the environment.