國立高雄大學圖資館 |

智慧型機器人應用基因演算法結合增強式學習於模糊控制歸屬函數之最佳化 = The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots

紀錄類型:	書目-語言資料,印刷品 : 單行本
並列題名:	The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots
作者:	林宜陞,
其他團體作者:	國立高雄大學
出版地:	[高雄市]
出版者:	撰者;
出版年:	2012[民101]
面頁冊數:	77面圖，表格 : 30公分;
標題:	智慧型機器人
標題:	intelligent robot
電子資源:	http://handle.ncl.edu.tw/11296/ndltd/18835223092246433315
附註:	104年10月31日公開
附註:	參考書目：面63-69
摘要註:	在本研究中，我們提出一種結合基因演算法(genetic algorithms)及增強式學習(reinforcement learning)的方法，進行模糊歸屬函數(fuzzy membership function)的最佳化，並以輪型機器人的運動控制驗證之。模糊控制是智慧型機器人常用的控制方式之一，其模糊歸屬函數的制定通常是由專家主觀(heuristic)認定，往往需花費大量的時間尋找最佳的組合，且不一定能適應各種環境。本研究透過機器學習(machine learning)的技術尋找能適應環境的模糊歸屬函數，達到最佳化機器人控制的目的。由於直接在機器人機體上進行在線式機器學習(on-line learning)會耗費許多時間，同時也對硬體機構造成很大的傷害；而離線式機器學習(off-line learning)採用模擬的方式進行學習，其假設往往過於理想，而使結果無法實用化。為了解決此問題，本研究提出兩階段的學習方法，第一階段以基因演算法進行離線學習，在模擬器中尋找理想狀態下的最佳模糊歸屬函數，在沒有硬體的損耗的情形下節省學習與演化的時間。第二階段將第一階段所獲得的最佳模糊歸屬函數，套用至實體機器人的控制系統，在現實環境下的回饋測試並以增強式學習法調整出最佳的模糊控制歸屬函數。由於第一階段的學習結果具有相當的可信度，第二階段的在線學習可以在較少的時間內學習出好的結果，可以降低硬體損耗同時縮減學習時間。本研究實際建立輪型服務型機器人，同時設計兩階段的機器學習演算法與控制機制，並實際測試其成效。實驗結果顯示本方法的確可以讓系統完成最佳化機器人模糊控制參數的調整。 Fuzzy control is one of the popular control methods for intelligent robots. Usually, the membership functions associated with the fuzzy control rules are defined by domain experts heuristically. It usually takes a lot of time to adjust the fuzzy membership functions before they can be used practically. Machine learning techniques are adopted in this study for the optimization of fuzzy membership functions. Typical machine learning methods can be divided into two classes, the on-line learning and the off-line learning. On-line learning methods perform on physical machines and practical environments, whereas off-line learning methods run in computer-simulated environments. For robot applications, thousands of on-line learning iterations may be needed that take a long learning time and may damage the robot hardware. The results from off-line learning methods may not be applicable on a real robot due to the ideal assumptions, but unreality, of the simulators. The study proposed a two-stage learning approach. In the first stage, the genetic algorithm is used to perform off-line learning for deriving the best membership functions in a simulator. In the second stage, the fuzzy membership functions obtained from the simulator are on-line adjusted on a real robot using the techniques of reinforcement learning. The learning outcomes from the first stage are not practical but reasonable and they can be adjusted with fewer iterations of on-line learning in the second stage. In this manner, the overall learning time can be reduced with less damage on the robot hardware. The proposed method was verified on a real, wheel-type vision-based, robot. The experimental results show that the proposed method effectively produce optimal fuzzy membership functions for the control of a real robot.

智慧型機器人應用基因演算法結合增強式學習於模糊控制歸屬函數之最佳化 = The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots
林, 宜陞

智慧型機器人應用基因演算法結合增強式學習於模糊控制歸屬函數之最佳化 = The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots / 林宜陞撰 - [高雄市] : 撰者, 2012[民101]. - 77面 ; 圖，表格 ; 30公分.
104年10月31日公開參考書目：面63-69.
智慧型機器人intelligent robot

智慧型機器人應用基因演算法結合增強式學習於模糊控制歸屬函數之最佳化 = The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots
LDR:05047nam0a2200289 450 001 346232
005 20170214095717.0
009 346232
010 0 $b 精裝
010 0 $b 平裝
100 $a 20170214d2012 k y0chiy50 e
101 0 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 智慧型機器人應用基因演算法結合增強式學習於模糊控制歸屬函數之最佳化 $d The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots $z eng $f 林宜陞撰
210 $a [高雄市] $c 撰者 $d 2012[民101]
215 0 $a 77面 $c 圖，表格 $d 30公分
300 $a 104年10月31日公開
300 $a 參考書目：面63-69
314 $a 指導教授：吳志宏教授
328 $a 碩士論文--國立高雄大學電機工程學系碩士班
330 $a 在本研究中，我們提出一種結合基因演算法(genetic algorithms)及增強式學習(reinforcement learning)的方法，進行模糊歸屬函數(fuzzy membership function)的最佳化，並以輪型機器人的運動控制驗證之。模糊控制是智慧型機器人常用的控制方式之一，其模糊歸屬函數的制定通常是由專家主觀(heuristic)認定，往往需花費大量的時間尋找最佳的組合，且不一定能適應各種環境。本研究透過機器學習(machine learning)的技術尋找能適應環境的模糊歸屬函數，達到最佳化機器人控制的目的。由於直接在機器人機體上進行在線式機器學習(on-line learning)會耗費許多時間，同時也對硬體機構造成很大的傷害；而離線式機器學習(off-line learning)採用模擬的方式進行學習，其假設往往過於理想，而使結果無法實用化。為了解決此問題，本研究提出兩階段的學習方法，第一階段以基因演算法進行離線學習，在模擬器中尋找理想狀態下的最佳模糊歸屬函數，在沒有硬體的損耗的情形下節省學習與演化的時間。第二階段將第一階段所獲得的最佳模糊歸屬函數，套用至實體機器人的控制系統，在現實環境下的回饋測試並以增強式學習法調整出最佳的模糊控制歸屬函數。由於第一階段的學習結果具有相當的可信度，第二階段的在線學習可以在較少的時間內學習出好的結果，可以降低硬體損耗同時縮減學習時間。本研究實際建立輪型服務型機器人，同時設計兩階段的機器學習演算法與控制機制，並實際測試其成效。實驗結果顯示本方法的確可以讓系統完成最佳化機器人模糊控制參數的調整。 Fuzzy control is one of the popular control methods for intelligent robots. Usually, the membership functions associated with the fuzzy control rules are defined by domain experts heuristically. It usually takes a lot of time to adjust the fuzzy membership functions before they can be used practically. Machine learning techniques are adopted in this study for the optimization of fuzzy membership functions. Typical machine learning methods can be divided into two classes, the on-line learning and the off-line learning. On-line learning methods perform on physical machines and practical environments, whereas off-line learning methods run in computer-simulated environments. For robot applications, thousands of on-line learning iterations may be needed that take a long learning time and may damage the robot hardware. The results from off-line learning methods may not be applicable on a real robot due to the ideal assumptions, but unreality, of the simulators. The study proposed a two-stage learning approach. In the first stage, the genetic algorithm is used to perform off-line learning for deriving the best membership functions in a simulator. In the second stage, the fuzzy membership functions obtained from the simulator are on-line adjusted on a real robot using the techniques of reinforcement learning. The learning outcomes from the first stage are not practical but reasonable and they can be adjusted with fewer iterations of on-line learning in the second stage. In this manner, the overall learning time can be reduced with less damage on the robot hardware. The proposed method was verified on a real, wheel-type vision-based, robot. The experimental results show that the proposed method effectively produce optimal fuzzy membership functions for the control of a real robot.
510 1 $a The Optimization of Fuzzy Membership Functions Using the Techniques of Genetic Algorithm and Reinforcement Learning for Vision-based Mobile Robots $z eng
610 0 $a 智慧型機器人 $a 機器學習 $a 模糊控制 $a 模糊歸屬函數 $a 基因演算法 $a 增強式學習
610 1 $a intelligent robot $a machine learning $a fuzzy control $a fuzzy membership functions $a genetic algorithm $a reinforcement- learning
681 $a 008M/0019 $b 542201 4437.1 $v 2007年版
700 1 $a 林 $b 宜陞 $4 撰 $3 576441
712 0 2 $a 國立高雄大學 $b 電機工程學系碩士班 $3 166118
801 0 $a tw $b NUK $c 20121108 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/18835223092246433315