臺灣博碩士論文加值系統

輪型機器人是近年來最常被使用的一種機器人移動方式，尤其是應用於農業的機器人系統，有別於雙足機器人，輪型機器人有較高的靈活性以及較低的技術門檻。而在農用機器人的應用中，搭載機械手臂具備輔助相關農事及運搬作業等功能之輪型機器人，一直是解決勞力不足甚至邁向無人化生產極富潛力的技術選項。因此本研究將以實驗室前輩所研發之R7-M2 I型機器人為雛形，除進行機構改良外，並置入Human-Following功能，期望提升機器人自主跟隨以及負載循跡之能力，以滿足各式農業生產模式、降低農民勞動力負荷並提升工作效率之性能需求。
本研究使用雷射測距儀及三軸複合羅盤做為回授之訊號源，並經由LabVIEW軟體將回傳之資料做紀錄與分析，再透過傳統開關控制器、模糊控制器以及基因模糊控制器等三種控制模式引導機器人執行跟隨行動並進行比較。成效之驗證則規劃設計無障礙物路線實驗與有障礙物路線實驗各15組共30組路線實驗，以比較在不同環境條件下和不同控制策略下跟隨成效之優劣。跟隨成效之判定以(1)單顆馬達完成任務消耗之總激磁步數(做為能源消耗之參考指標)(2)步進馬達轉速之變異係數(做為機器人移動之穩定性指標)(3)完成任務時機器人與跟隨目標之位移(做為跟隨任務完成度之指標)三項指標進行評比。
經過30組之不同路線與3種不同的控制策略共90組實驗結果顯示，應用基因模糊控制器於輪型移動機器人進行跟隨行為控制之可行性獲得驗證，而基因模糊控制器相較於前述其他二種控制法則，在輪型機器人跟隨控制之穩定性以及跟隨任務完成之準確性，皆展現較佳之成效，為三種控制法之首選。

In recent years, wheeled mobile robots are the most popular mobile robots, especially in agricultural field. As a wheeled agricultural mobile robot, it is said to be more flexible and easier-controlled when compared with biped robots. When the robot arms mounted, wheeled mobile robots are supposed to be the solutions of labor-lack or unmanned farm. Therefore, in this research, based on robot R7-M2 I which was built by senior researcher Luo, a human-following algorithm was developed and integrated into the reconstructed R7-M2 I as robot L1-M4 II, that is expected to reduce the heavy burdens but promote the efficiency for farm works simultaneously. A scanning laser rangefinder and one three-axis gyroscope were configured to be feedback signal sources in this thesis. With LabVIEW codes, the feedback signals were recorded and analyzed as the reference signals of the on-off controller, the fuzzy logic controller and the genetic fuzzy logic controller, the three types of competitive controllers, for establishing human following behavior. 15sets of obstacle free and obstacle-included human-following path experiments were conducted to evaluate the validity and performance respectively. The validity of human-following was assessed numerically by total step counts of every step motor (served as the index of power cost), C.V of velocity of each step motor (served as index of stability of robot movement) and mission-completed displacement (served as index of effectiveness of robot movement). The results induced from 90 experiments show that the human following behavior controlled by a genetic fuzzy logic controller for wheeled mobile robots is feasible and at the same time the genetic fuzzy logic controller, with characteristics of more stability, better human-following performance, but less power cost, is the best choice among the three competitive controllers.

目錄 v
表目錄 viii
第一章 序論 1
1.1 前言 1
1.2 研究目的 3
第二章 文獻探討 5
2.1 雷射測距儀的原理與特性 6
2.2 模糊理論 8
2.3 基因演算法 11
2.3.1 基因演算法理論 12
2.3.2基因演算法文獻探討 15
2.4基因模糊控制理論 17
2.4.1基因演算法融合模糊理論 17
2.4.2基因模糊文獻探討 18
第三章 實驗設備 20
3.1 實驗器材 20
3.1.1控制元件Arduino 20
3.1.2三軸複合羅盤 22
3.1.3 雷射測距儀 24
3.1.4 步進馬達 27
3.1.5 馬達驅動模組 28
3.1.6 穩壓晶片 31
3.2 L1-M4 II型機器人 34
3.3 LabVIEW 35
第四章 控制策略設計 37
4.1 控制流程 37
4.2 模糊控制轉彎 38
4.3 模糊控制避障 40
4.4 直線跟隨控制策略 42
4.4.1 直線跟隨---傳統控制法 42
4.4.2 直線跟隨---模糊控制法 42
4.4.3 直線跟隨---基因模糊控制法 44
4.4.3.1 訓練模式概述 44
4.4.3.2 訓練模式細部設計 47
4.4.3.3 執行模式 51
第五章 實驗方法 52
5.1實驗路線設計 53
5.2 控制策略 56
5.3 成效判別依據 56
第六章 實驗結果與討論 58
6.1 路線規劃及實驗範例 59
6.2 第1~15組實驗結果 73
6.3 第16~30組實驗結果 76
6.4 第1~30組實驗結果綜合分析與討論 78
第七章 結論與建議 80
7.1 結論 80
7.2 建議 80
參考文獻 82

[1] 蘇木春、張孝德。2000。機器學習類神經網路、模糊系統以及基
因演算法則。台北：全華圖書公司。
[2] 羅茂榮。”模糊理論應用於自主性機器人避障路徑導引之研
究”台中：國立中興大學生物產業機電工程學系碩士論文。 2014。
[3] 林昇甫、徐永吉。2009。遺傳演算法及其應用。台北：五南圖書
公司。
[4] 莊謙本、周明、蕭培墉、蘇崇彥。2013。智慧型機器人技術與應
用。台北:全華圖書。
[5] W. Chung, H. Kim, Y. Yoo, C.B. Moon, and J. Park, ”The
detection and following of human legs through inductive
approaches for a mobile robot with a single laser range
finder” IEEE Trans.Ind.Electron. vol.59 , no.8 ,
pp.3156-3165 , AUG. 2012.
[6] J. S. Hu, J. J. Wang, and D.M. Ho, ”Design of sensing
system and anticipative behavior for human following of
mobile robots” IEEE Trans. Ind. Electron . vol.61 ,
no.4, pp.1916-1925, APRIL 2014.
[7] R.W. Deng, Y. H. Wang , C.J Lin, and T.H.S. Li, ”Implementation of human following mission by using fuzzy
head motion control and Q-Learning wheel motion control
for home service robot” International Conference on
Fuzzy Theory and Its Application, Dec.6-8, 2013, pp53-60.
[8] W. J. Wang, J. W. Chang,” Implementation of a mobile
robot for people following,” International Conference
on System Science and Engineering , June 30-July 2, 2012,
pp112-116.
[9] M. y. Zhu, and Y. l. Y ang, ”A fuzzy fusion algorithm
for radar/infrared sensors,” Radar Conference, 2009 IET
International, Beijing Institute of Technology, Beijing
100081, April 2009 , pp1-4.
[10] A. Mandow, ”the autonomous mobile robot aurora for
greenhouse operation,” IEEE Robotics & Automation
Magazine, DEC.1996, pp18-27.
[11] A. Fernando and C. Ricardo, ”Agricultural robotics-
unmanned robotic service units in agricultural tasks,”
IEEE Industrial Electronics Magazine , SEP.2013, pp. 49-
57.

[12] F. F.AM, Z. Nakaot, Y. W. Chen, ”Playing the rock-
paper-scissors game with a genetic algorithm,” IEEE,
pp741-745, 2000.
[13] S. Jiang, J. S. Cheng , J. P. Chen ,”Design of central
pattern generator for humanoid robot walking based on
multi-objective GA,” Proceedings of the 2000 IEEE/RSj
international Conference on Intelligent Robots and
Systems, pp1930-1935.
[14] X. Lu, J. Yi, ”Autopilot design based on fuzzylogic
and genetic algorithm,” IEEE Institute of Automation,
Chinese Academy of Sciences, 2006.
[15] C. Guo, J. Hu and T. Li, ”Design of GA-based fuzzy
control for ship steering,” Proceedings of the 6th World
Congress on Intelligent Control and Automation, June 21 -
23, 2006, pp3594-3598.
[16] M. Y. Ju and C. W. Cheng, ”Smooth path planning using
genetic algorithms,” Proceedings of the 8th World
Congress on Intelligent Control and Automation, June 21-
25 ,2011, pp1103-1107.

[17] Honeywell cop. ”Three-axis compass with algorithms
HMC6343,” Internet:
http://www51.honeywell.com/aero/common/documents/myaerospac
ecatalog-documents/Missiles-Munitions/HMC6343.pdf , May
2014.
[18] Hokuyo Automatic co., LTD. “Scanning laser range
finder URG-04LX-UG01(Simple-URG),“Internet:
http://www.hokuyo-
aut.jp/02sensor/07scanner/download/pdf/URG-
04LX_UG01_spec_en.pdf, AUG.2009.
[19] T. B. Alamgir , I. J. Alfar , M.M. Rashid , “Modeling
and robust analysis of a fuzzy based person following
robot,” International Conference on Computer and
Communication Engineering , IEEE , July,3-5, 2012, pp358-
361.
[20] 維基百科 , 條目: Kinect.
[21] H. J. Cho , B.H. Wang , S. Roychowdhury , “Automatic
rule generation for fuzzy controllers using genetic
algorithms: A study on representation scheme and mutation
rate, ” IEEE ,1998., pp1290-1295.
[22]U. Ansari , S. Alam , “Hybrid genetic algorithm fuzzy
rule based guidance and control for launch vehicle,”
2011 11th International Conference on Intelligent Systems
Design and Applications , IEEE, 2011 , pp178-185.
[23]STMicroelectronics , “L298N dual full-bridge driver,”
Internet: www.datasheetcatalog.com
[24]鑫奇智慧科技89idea.com, “Arduino L298N 2A 雙路直流馬達
驅動板,” Internet: http://www.89idea.com/blog/?p=132
,Sep.6,2011.
[25]喬治查爾斯電子電路網, ”LM317/LM337穩壓電路,” Internet:

http://georgecharles.idv.st, May26, 2016.
[26]Texas Instrument, ”LM117,LM317-N Wide temperature
three-pin adjustable regulator,” Oct. 2015 .
[27]K. Morioka, J.H. Lee, H.Hashimoto, “Human-Following
mobile robot
in a distributed intelligent sensor network,” IEEE
Transactions On
Industrial Electronics, vol.51, no1, pp229-237, Feb.2004 .