室內居家服務機器人之研究; The Study of an Indoor Service Robot System

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/9823

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/9823

Title:	室內居家服務機器人之研究;The Study of an Indoor Service Robot System
Authors:	丁穆仁;Mu-Jen Ting
Contributors:	資訊工程研究所
Keywords:	導航;定位;居家服務;機器人;模糊控制;攝影機;indoor service;robot;fuzzy control;camera;location;navigation
Date:	2009-07-06
Issue Date:	2009-09-22 11:57:06 (UTC+8)
Publisher:	國立中央大學圖書館
Abstract:	台灣已經成為全球高齡化國家之一，因觀念習俗使得老人家不願意至療養院集中照顧，反而選擇與子女同居，平日子女外出工作，家裡只有老人時，將成為老人發生危險程度最高的時段。有鑒於目前的遠距居家看護系統所提供的服務，疑似發生意外時以文字警告訊息為主，較為缺乏即時的現場狀況資訊。因此本論文期望能研究設計一套居家看護系統，以機器人為主要的服務人機介面，在疑似發生意外狀況時，快速提供現場近距的離影像資訊。本論文使用市售的機器人平台為介面，為機器人設計模糊控制系統，透過無線網路控制機器人，結合建築物內已架設的室內監視系統，以攝影機為主要追蹤以及導航機器人的工具。而室內攝影機多以斜向監視室內環境，因為有角度而需要做修正，減少使用者架設的複雜度，本論文使用統計的方法，僅擷取少量影像資訊，使用三次回歸曲線來推估整個環境的影像變化，以確實追蹤機器人位置。而此推估曲線也將利用在，修正機器人因高度的影響，判斷其在地板上的位置，還有模糊控制時要輸入的角度與距離。用加速度計結合無線射頻收發模組，來感應老人是否發生跌倒狀況，或是門口是否有人開門，當電腦判定意外發生時啟動機器人執行服務。最後本論文以系館內的辦公室，模擬一般家庭室內情境，做幾個不同的實驗來測試本系統的效能與穩定度，證明本系統是可以完成居家服務的效果，而且具有預期中的便利性、擴充性的系統架構。 There are more old people in Taiwan now. Because of the Chinese custom, elder person lives at home with their children rather than in sanatorium. It is very danger when all family out of home in the daytime. We worry about the distance home service only send character message when an accident happened. The message lacks for some view of the situation in real time. For this reason, this paper went to design an indoor service system to get the view for worry fast. The system uses a robot for human-computer interface. We use the commercial robot platform to be an interface, design a fuzzy control system for the robot, and control it through the wireless network in this paper. Combining the indoor monitoring system erected in the building, we use the camera as the main tracking and navigating tool. Consider about revising the camera with oblique, we use the method of statistics and only need a few images in order to reduce the complexity of erecting this system. It used the three order polynomial function to estimate the image change to track the robot. The results of estimating also make use of revising the height of robot and the input of fuzzy system. The system watch whether old man fallen or someone opened door with accelerometer RF-module. Computer will start the robot to serve when it judges the accident takes place. In the end, we simulate a general family’s indoor situation with office in CSIE building, and do several different experiments to test this system’s stability and efficiency. This paper proves the indoor service robot is workable, convenient and expandable.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Size	Format

社群 sharing

Loading...