淡江大學機構典藏:Item 987654321/35805
English  |  正體中文  |  简体中文  |  全文笔数/总笔数 : 62797/95867 (66%)
造访人次 : 3730529      在线人数 : 627
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜寻范围 查询小技巧:
  • 您可在西文检索词汇前后加上"双引号",以获取较精准的检索结果
  • 若欲以作者姓名搜寻,建议至进阶搜寻限定作者字段,可获得较完整数据
  • 进阶搜寻


    jsp.display-item.identifier=請使用永久網址來引用或連結此文件: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/35805


    题名: 智慧型娛樂機器人之研製
    其它题名: Design and implementation of intelligent entertainment robots
    作者: 鄭吉泰;Cheng, Chi-tai
    贡献者: 淡江大學電機工程學系博士班
    翁慶昌;Wong, Ching-chang
    关键词: 機器手臂;機器寵物;人形機器人;機器學習;Robot arm;Pet robot;humanoid robot;Machine learning
    日期: 2008
    上传时间: 2010-01-11 07:11:06 (UTC+8)
    摘要: 本論文研製三台智慧型娛樂機器人來分別實現三項互動:三維空間互動、情緒感覺互動與娛樂競賽互動。首先,本論文設計與實現一台具有雙手臂、雙輪與雙視覺系統的居家機器人,其可以在三維空間裡抓取目標物來做三維空間互動。機器人上半身裝置四個自由度的雙臂,並且在頭部與手部各裝設一個視訊裝置。本論文應用一個類神經架構來縮小機器人視覺上的誤差,讓所實現的機器人能以雙視覺搭配手臂與兩輪在三維空間中抓取特定目標物。其次,本論文設計與實現一台具有四足、兩個耳朵、一個嘴巴與一個視覺系統的四足機器寵物狗,其可以依照人類的互動情況產生不同的情緒反應來做情緒感覺互動。本論文提出一個可以自我行為學習的類神經學習架構,讓所實現的機器狗能在跟人與環境的互動後,自我學習到可以依照不同環境情況表現出不同的情緒反應。最後,本論文設計與實現一台具有二手、二足與一個視覺系統的人形機器人,其可以實現踢足球等娛樂競賽行為。26個自由度的機構設計讓讓所實現的人形機器人可以完成前進、後退、旋轉及側移等動作。頭部、手部、腰部及腿部分別為2個、8個、2個及14個自由度。ㄧ個視訊裝置裝設在機器人之頭部,讓其可以為一台視覺自主的機器人。本論文根據機器人身上指南針與紅外線感測器所獲得之訊息提出一個模糊避障演算法,其讓機器人可以自主有效的避開障礙物並且快速的走到目標位置。從模擬結果與實體的實際驗證可知,所完成之兩輪雙視覺機器人、四足機器寵物狗及二足人形機器人分別在三維空間互動、情感互動與娛樂競賽互動均有不錯的性能。
    Three intelligent entertainment robots are designed and implemented for three kinds of interaction. First, a home robot with two arms, two wheels, and two vision systems is design to catch a specific target in a three-dimensional space so that it can do some three-dimensional physical interactions. Two CMOS sensors are respectively mounted on the head and the right arm and the arm with 4 degrees of freedom (DOF) is design. A neural network learning system is proposed and built on the robot so that it is able to use its vision systems to track the object and use its right arm to catch the object. Second, a pet robot with four legs, two ears, one mouth and one vision system is design to do different emotional interaction in different situation so that it can do some emotional interaction with human. A neural network learning system is proposed and built on the robot for the behavior learning so that the implemented pet robot is able to choose a suitable behavior for different environmental situation. Third, a humanoid robot with two arms, two legs, and one vision system is design to play soccer so that it can do some entertainment and competition interaction. A mechanical structure with 26 DOF is designed so that the implemented humanoid robot can find the ball, walk forward, turn right and left, and slip right and left. There are 2 DOF on the neck, 2 DOF on the waist and trunk, 8 DOF on the two arms, and 14 DOF on the two legs. One vision system is mounted on the head of robot so that it is a vision-based soccer robot. A fuzzy obstacle avoidance method based on the obtained information of six infrared sensors and one electronic compass is proposed and implemented on the humanoid robot so that it can avoid obstacles effectively and go to the destination area quickly. From the simulation and experiment results, we can see that the implemented three robots have good performance in the three-dimensional physical interaction, emotional interaction, and entertainment interaction.
    显示于类别:[電機工程學系暨研究所] 學位論文

    文件中的档案:

    档案 大小格式浏览次数
    index.html0KbHTML316检视/开启

    在機構典藏中所有的数据项都受到原著作权保护.

    TAIR相关文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回馈