![]() |
|
以作者查詢圖書館館藏 、以作者查詢臺灣博碩士 、以作者查詢全國書目 、勘誤回報 、線上人數:8 、訪客IP:3.129.26.177
姓名 陳易鋌(Yi-Ting Chen) 查詢紙本館藏 畢業系所 電機工程學系 論文名稱 具模仿學習與旋律辨識之打擊樂器機器人
(The percussion robot with music learning and melody recognition)相關論文 檔案 [Endnote RIS 格式]
[Bibtex 格式]
[相關文章]
[文章引用]
[完整記錄]
[館藏目錄]
[檢視]
[下載]
- 本電子論文使用權限為同意立即開放。
- 已達開放權限電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
- 請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
摘要(中) 本論文目標為設計一具有聽音、辨音與演奏功能之機器人,經由使用者與機器人在音樂上的互動,來達到娛樂與教育的效果。此機器人使用七顆AX-12馬達來組成雙手與腰部,每隻手持有一根鼓棒,利用腰部轉動再配合雙手開合以及打擊的動作,來完成演奏的功能。此機器人也具聽音、辨音功能,以麥克風收音與電腦端分析來執行之。機器人與使用者互動中,包含兩主要模式:模式一為「模仿模式」,使用者可隨意在電子琴演奏一段音樂,機器人可以聽音、記憶並加以辨識,模仿出使用者彈奏的旋律;模式二為「指定曲目模式」,我們建立了30首歌曲樂譜的資料庫,機器人可以實現以下兩種表演:1. 使用者任選資料庫中之某一首曲子,並用電子琴任意彈奏某連續兩小節,機器人亦能正確辨識出是哪一首曲子,並表演出完整全曲。2. 若使用者彈錯某些音符或變調(如:C大調錯彈成G大調),機器人仍能正確辨識出是哪一首曲子,並雙部演奏全曲。本論文之成果可應用於協助初學音樂的孩童或成人,透過與機器人互動增加學習音樂之興趣,進而學習樂理基本認識與聽力練習,兼具寓教於樂的效果。
摘要(英) The goal of this thesis is to design a robot which has interaction with human by performing music learning and playing. This study can be an application for both education and recreation. The robot not only can listen, recognize and memorize the melody, but also can play the recognized music. The robot has a waist and two hands composed of seven AX-12 motors. Each hand holds one drumstick. Combining two hands’ beating motion and waist rotation, the robot can play music in the music bricks. The interaction between the robot and human has two modes. Mode 1:The user plays a music in an electric keyboard, then the robot listens and recognizes the melody. Finally the robot plays the melody that was played by the user. In Mode 2, we have thirty songs available for the user to choose. The user only requires to play at least eight notes of the melody from any part of the song. The robot recognizes the pitch and the rhythm of the melody, then it will choose the correct song to play.
關鍵字(中) ★ 旋律辨識
★ 音訊處理
★ 打擊樂器機器人關鍵字(英) ★ audio processing
★ percussion robot
★ melody recognition論文目次 摘要.........................................i
Abstract....................................ii
誌謝.......................................iii
目錄........................................iv
圖目錄......................................vi
表目錄......................................ix
第一章緒論...................................1
1.1研究背景與動機............................1
1.2論文目標..................................2
1.3本文架構..................................2
第二章系統架構與實驗環境.....................4
2.1機器人機構設計............................6
2.2系統硬體介紹.............................10
2.3軟體架構.................................12
2.4實驗環境介紹.............................13
第三章音訊辨識..............................15
3.1音訊資料擷取與分析.......................17
3.2音高辨識.................................19
3.2.1特徵選取與樣板建立.....................19
3.2.2樣板比對...............................20
3.2.3相似度計算方法.........................22
3.3歌曲旋律辨識.............................24
3.3.1樣板比對...............................25
3.3.2自選情況...............................27
3.3.3失誤情況與特殊案例.....................28
3.3.4變調情況...............................30
第四章打擊路徑規劃與控制策略................32
4.1打擊路徑規劃.............................32
4.1.1組成路徑...............................33
4.1.2規劃條件...............................34
4.2路徑篩選.................................36
4.2.1路徑分析與淘汰.........................37
4.2.2路徑比較...............................39
4.3延遲時間控制策略.........................41
4.3.1輸入與輸出.............................42
4.3.2歸屬函數...............................43
4.3.3模糊規則庫.............................44
4.3.4模糊推論與解模糊化.....................45
4.4馬達控制封包格式.........................46
第五章實驗成果..............................48
5.1模仿模式.................................48
5.2指定曲目模式.............................50
第六章結論與未來展望........................60
6.1結論.....................................60
6.2未來展望.................................60
參考文獻....................................62
參考文獻 [1]N. Kohl and P. Stone, “Policy gradient reinforcement learning for fast quadrupedal locomotion,” The Proceedings of IEEE International Conference on Robotics and Automation, New Orleans, LA, April 26-May 1, pp.2619-2624, 2004.
[2]介紹AIBO之Sony公司網站 http://www.sony.net/SonyInfo/News/Press_Archive/200105/01-027E/
[3]T. S. Perry, “Pleo, the Poop-Free Pet,” IEEE Spectrum, vol. 45, pp.24-24, 2008.
[4]介紹Pleo之UGOBE公司網站 http://www.pleoworld.com/
[5]SEGA公司寵物機器人產品網站 http://www.idog-segatoys.com/
[6]H. Kozima, C. Nakaqawa and Y. Yasuda “Designing and observing human-robot interactions for the study of social development and its disorders,” The Proceedings of IEEE International Conference on Computational Intelligence in Robotics and Automation, Espoo, Finland, June 27-30, pp.41-46, 2005.
[7]介紹Keepon之網站 http://www.myu.ac.jp/~xkozima/index.html
[8]Y. Nonomura, M. Fujiyoshi and H. Sugihara “Inertial Force Sensing System for Partner Robots,” The Proceedings of IEEE International Conference on Sensors, Exco, Daegu, Korea, October 22-25, pp.1325-1328, 2007.
[9]L. Geppert, “Qrio, the robot that could,” IEEE Spectrum, vol. 41, pp.34-37, 2004.
[10]G. Endo, J. Nakanishi, J. Morimoto and G. Cheng, “Experimental studies of a neural oscillator for biped locomotion with QRIO,” The Proceedings of IEEE International Conference on Robotics and Automation, Barcelona, Spain, April 18-22, pp.596-602, 2005.
[11]B. Tribelhorn and Z. Dodds, “Evaluating the Roomba: A low-cost, ubiquitous platform for robotics research and education,” The Proceedings of IEEE International Conference on Robotics and Automation, Roma, Italy, April 10-14, pp.1393-1399, 2007.
[12]Lawn Mowing Robot介紹網站 http://www.robotlawnmowers.ie/index.html
[13]王小川 編著, 語音訊號處理, 全華圖書股份有限公司, 2007.
[14]鄭琇云 (張智星教授 指導) , “各種音高追蹤方法對哼唱選歌之影響的評估,” 碩士論文, 清華大學資訊工程所, 2008.
[15]李念容 (張智星教授 指導) , “哼唱檢索的辨識方法改進及探討,” 碩士論文, 清華大學資訊工程所, 2007.
[16]優勢音樂生活網http://www.yusmusic.com/index.php?act=product&CategoryID=98
[17]王文俊 編著, 認識Fuzzy-第三版, 全華圖書股份有限公司, 2007.
[18]智慧型控制與影像處理實驗室網站http://140.115.70.172/xoops/modules/wfdownloads/singlefile.php?cid=3&lid=36
指導教授 王文俊(Wen-June Wang) 審核日期 2009-7-7 推文 plurk
funp
live
udn
HD
myshare
netvibes
friend
youpush
delicious
baidu
網路書籤 Google bookmarks
del.icio.us
hemidemi
myshare