臺灣博碩士論文加值系統

受到疫情嚴重衝擊的背景以及人工智慧的新興發展之下，自動化技術已逐漸滲透生活與工作之中。從智能家居的自動化控制系統，到工廠的無人自動化生產線，甚至於目前推廣的智慧農業，以上敘述皆需較長的架設時間，並且在多樣性和通用性上還存在限制。因此，本論文將專注於如何開發出具有多樣性功能並具有廣泛通用性的自動化技術，透過開發一款自主移動機器人(Autonomous Mobile Robot, AMR)，致力於開發一款不僅能在室內與戶外進行導航及避障，並且配備了多種功能設備，以便應用在各種情境中提升其實際使用價值。為達到此目標，將採用多個機器人作業系統(Robot Operating System, ROS)功能包，以實現機器人室內外的導航及避障，機器人上安裝自動噴灑模組及協作型機械手臂。本研究將工作區域分為室內實驗和戶外實驗，以驗證其多功能室內/戶外適應性的目標。在室內實驗中，將採用光學雷達（Lidar）來收集掃描數據，通過Gmapping演算法進行地圖建置，完成地圖即可透過move_base功能包進行路徑規劃，因自主移動機器人導航到點位仍會有些許的誤差，導致無法執行上下料工作任務。因此，在機械手臂上加裝相機及夾爪，並在各工作站貼取標籤，透過Aruco可得知鏡頭到標籤的六個自由座標，即可推算出標籤、鏡頭、夾爪及機械手臂之間的空間，以實現上下料工作任務，在室內中開發一套自主設計的主控制系統，使機器人能夠自行規劃路徑以及執行上下料的工作。在戶外實驗中，在move_base功能包中導入全球定位系統（Global Positioning System, GPS）模組，以實現機器人在戶外環境導航中得知機器人本身的經緯度，並可透過下達工作點位的經緯度進行導航。在機械手臂的應用部分，將導入YOLOv4物體辨識技術。不僅能方便地獲得目標物體的座標位置，經過座標轉換後，機械手臂就可以進行精確的後續動作。這種方法可以有效地解決戶外環境中變數眾多的問題，避免物體辨識的難題，確保的機器人在各種環境條件下都能適應並有效地執行任務。此外，在車體兩側安裝雙軸運動平台，以實現戶外噴灑的功能。透過此兩實驗能夠具體實現開發出具有多樣性功能並具有廣泛通用性的自動化技術的初衷。

Against the backdrop of severe impacts from the pandemic and the emerging development of artificial intelligence. From unmanned automated production lines in factories, and even to the currently promoted smart agriculture, all of these require lengthy setup times and still have limitations in diversity and versatility. This paper explores the development of versatile, universally applicable automation technology using an Autonomous Mobile Robot (AMR). We employ multiple Robot Operating System (ROS) packages for these functionalities, along with an automatic spray module and a collaborative robotic arm. In the indoor experiment, we use Lidar to collect scanning data and construct a map through the Gmapping algorithm. After completing the map, we use the move_base package for path planning. Through Aruco, we can obtain the three-dimensional coordinates of the lens to the tag, which can be used to calculate the space between the tag, lens, gripper, and robotic arm to achieve the task of loading and unloading. In the indoors, we developed a self-designed main control system so that the robot can plan its path and execute the loading and unloading work. Outdoors, we integrate a GPS module for navigation and use YOLOv4 object recognition for accurate task execution, combating outdoor environment variables. The AMR is also equipped with a dual-axis motion platform for outdoor spraying. Through these experiments, we strive to advance diverse, applicable automation technology.

摘要...................................i
Abstract..............................ii
目錄.................................iii
誌謝..................................iv
表目錄................................vi
圖目錄...............................vii
符號說明..............................ix
第一章 緒論...........................1
1.1 研究背景.......................1
1.2 文獻探討.......................2
1.2.1 戶外自動化機器人................2
1.2.2 AGV、AMR及AMMR.................3
1.2.3 機器視覺.......................4
1.3 機器人架構與開發環境............6
1.4 論文架構.......................8
第二章 自主移動機器人底盤定位及導航.....9
2.1 系統架構.......................9
2.2 車體底盤與控制架構..............10
2.3 地圖建置與即時定位..............14
2.3.1 距離感測器.....................14
2.3.2 地圖建構.......................15
2.4 自動導航及避障..................18
2.4.1 自適應蒙特卡羅定位法.............19
2.4.2 代價地圖(Costmap)..............20
2.4.3 路徑規劃.......................20
2.5 相機姿態評估系統................21
第三章 工作模組........................22
3.1 系統架構........................22
3.2 自動噴灑模組....................23
3.2.1 Dynamixel智能馬達...............23
3.2.2 智能馬達控制器..................24
3.3 物件夾取模組....................25
3.3.1 機械手臂........................25
3.3.2 夾爪控制........................26
3.4 YOLO視覺模組....................27
第四章 實驗結果........................29
4.1 室內搬運物料實驗.................30
4.1.1 物件夾取模組.....................31
4.1.2 室內整體動作流程.................34
4.2 戶外導航YOLO辨識實驗.............37
4.2.1 整體控制流程.....................38
4.2.2 GPS座標自動導航及避障實驗.........39
4.2.3 修剪細枝系統.....................41
4.2.4 自動噴灑平台.....................44
4.2.5 戶外整體動作流程..................46
第五章 結論............................48
參考文獻.................................49
附錄一...................................52
附錄二...................................53
Extended Abstract.......................57

[1]CES未來科技正發生/從自動化到智能化機器人、電動車再進化，https://udn.com/news/story/6905/6979178
[2]Robotics is making a comeback in agriculture，http://finance.people.com.cn/BIG5/n1/2022/0714/c1004-32474959.html
[3]Z. Dong, H. Wang, Y. Tang, M. Wang, J. Zhao (2019). Research on Automatic Driving System of Agricultural Machinery Based on Embedded. 2019 IEEE 3rd Information Technology, Networking,Electronic and Automation Control Conference (ITNEC 2019)1940-1943
[4]全地形農作探勘車, https://www.robotnik.eu/vinbot/
[5]H. Igawa, T. Tanaka, S. Kaneko (2009). Visual and Tactual Recognition of Trunk of Grape for Weeding Robot in Vineyards. 2009 35th Annual Conference of IEEE Industrial Electronics7274-7289
[6]草莓採收機, https://harvestcroo.com/about/
[7]Bug Vacuum, https://www.agrobot.com/bugvac
[8]K. Kim, A. Deb, and D. J. Cappelleri, "P-AgBot: In-Row & Under-Canopy Agricultural Robot for Monitoring and Physical Sampling," in IEEE Robotics and Automation Letters, vol. 7, no. 3, pp. 7942-7949, July 2022, doi: 10.1109/LRA.2022.3187275.
[9]Pellenc Optimum 890, https://www.hydralada.com/nz/products/viniculture/pellenc-multifunction-optimum-harvester-890/
[10]Apple harvester, htps://www.washingtonpost.com/news/national/wp/2019/02/17/feature/inside-the-race-to-replace-farmworkers-with-robots/?noredirect=on
[11]Grapevine Exploration Vehicle, http://www.grape-project.eu/results/
[12]Sweeper, https://buzzorange.com/techorange/2018/12/17/automatic-harvest-robot/
[13]自主移動機器人聯盟發布首份移動機器人安全標準提升移動機器人產品價值, https://www.itri.org.tw/ListStyle.aspx?DisplayStyle=01_content&SiteID=1&MmmID=1036276263153520257&MGID=111082610554702043
[14]D. Bore, A. Rana, N. Kolhare, U. Shinde (2019). Automated Guided Vehicle Using Robot Operating Systems. 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI) 816-822
[15]D. C. Conner, J. Willis (2017). Flexible Navigation: Finite state machine-based integrated navigation and control for ROS-enabled robots. SoutheastCon 2017
[16]M. Filipenko, I. Afanasyev (2018). Comparison of Various SLAM Systems for Mobile Robot in an Indoor Environment. 2018 International Conference on Intelligent Systems (IS) 400-407
[17]A. L. Ibáñez, R. Qiu, D. Li (2017). A Simple, Cost-Effective, and Practical Implementation of SLAM Using ROS and Arduino. 2017 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom), and IEEE Smart Data (SmartData) 835-840
[18]M. Tarek, A. Moataz, M. Khaled, A. Hammam, O. M. Shehata, E. I. Morgan (2018). Multisensor Filtration and Fusion on a Three-Layer Architecture. 2018 IEEE International Conference on Vehicular Electronics and Safety (ICVES)
[19]A. Yilmaz, H. Temeltas (2020). ROS Architecture for Indoor Localization of Smart-AGVs Based on SA-MCL Algorithm. Bursa, Turkey, Turkey
[20]M. Reke, D. Peter, J. S. Tigges, S. Schiffer, A. Ferrein, T. Walter, D. Matheis (2020). A Self-Driving Car Architecture in ROS2. Cape Town, South Africa, South Africa
[21]M. Pleune, N. Paul, C. Faulkner, C. Chung (2020). Specifying Route Behaviors of Self-Driving Vehicles in ROS Using Lua Scripting Language with Web Interface. Chicago, IL, USA, USA
[22]Tian, K. and K. Mirza (2022). Sensor Fusion for Octagon – an Indoor and Outdoor Autonomous Mobile Robot. 2022 IEEE International Systems Conference (SysCon).
[23]Dickens, J., et al. (2021). Development of a multipurpose, outdoor autonomous ground vehicle for agricultural inspection. 2021 Rapid Product Development Association of South Africa - Robotics and Mechatronics - Pattern Recognition Association of South Africa (RAPDASA-RobMech-PRASA).

[24]自主移動協作機器人 (AMMR) 的全球市場, 2028, https://www.gii.tw/report/marf1093765-autonomous-mobile-manipulator-robots-ammr-market.html
[25]cooperative car, https://www.airitilibrary.com/Publication/alDetailedMesh?docid=U0046-2908201814104400
[26]邁向智慧製造：機械手臂整合視覺的現況與未來, 2020,https://www.smartmolding.com/21-01a08/
[27]Automatic pruning of vines, https://www.visionrobotics.com/vr-grapevine-pruner
[28]J. Xiong, Z. Liu, R. Lin, R. Bu, Z. He, Z. Yang, and C. Liang (2018). “Green Grape Detection and Picking-Point Calculation in a Night-Time Natural Environment Using a Charge-Coupled Device (CCD) Vision Sensor with Artificial Illumination”, Sensors , Vol.18. no.4, 969
[29]L. Zechuan, W. Song. (2019). Broken Corn Detection Based on an Adjusted YOLO With Focal Loss. IEEE Access, 7, 68281-68289. doi: 10.1109/ACCESS.2019.2916842
[30]H. Yang, C. Long, C. Miaoting, M. Zhibin, D. Fang, L. Maozhen. (2019). Tender Tea Shoots Recognition and Positioning for Picking Robot Using Improved YOLO-V3 Model. IEEE Access, 7, 180998-181011. doi: 10.1109/ACCESS.2019.2958614
[31]Y. Yang, Z. Kailiang, L. Hui, Y. Li, Z. Dongxing. (2020). Real-Time Visual Localization of the Picking Points for a Ridge-Planting Strawberry Harvesting Robot. IEEE Access, 8, 116556-116568. doi: 10.1109/ACCESS.2020.3003034
[32]自動除草機器人, https://www.naio-technologies.com/en/agricultural-equipment/vineyard-weeding-robot/
[33]半自動農藥噴灑機器人, https://agronomybots.com/index.php/2017/09/27/semi-autonomous-vineyard-spraying-agricultural-robot-savsar-hellenic-open-university-experiments/
[34]太陽能除草機器人, https://www.vitirover.fr/en-innovation
[35]太陽能型農耕機器人, https://vitibot.fr/en
[36]大型農耕機器人, https://www.rovitisveneto.it/rovitis-4-0-presente-allevento-nova-in-vigneto-a-spresiano-tv/
[37]全地形農作探勘車, https://www.robotnik.eu/vinbot/
[38]C. C. Hung, H. Y. Hsu, W. C. Li, C. Y. Li, and J. D. Lee (2020). “Automatic collecting and shooting mobile robot,” international journal of mechanical engineering and robotics research, vol. 9, no. 3, pp. 401-407
[39]J. D. Lee, C. Y. Li, K. W. Chen, and S. Y. Chen (2020). “ROS-based automated guided electric vehicle development”, International Automatic Control Conference CACS 2020, #1102
[40]李政毅，2020，”自動配送機器人研製”，國立虎尾科技大學自動化工程系碩士論文，https://hdl.handle.net/11296/453phh
[41]陳冠瑋，2021，”多自動導引車與工業機械手臂協同作業系統開發與實現”， 國立虎尾科技大學自動化工程系碩士論文，https://hdl.handle.net/11296/ce4y6v
[42]Chun-Tse Lee, Jen-Yuan Chang(2021), 光學視覺與機械手臂系統整合之校正方法介紹, https://www.tiri.narl.org.tw/Files/Doc/Publication/InstTdy/226/02260243.pdf
[43]S. Thrun, W. Burgard, and D. Fox, “Probabilistic robotics,” Cambridge, MA, USA: MIT Press, 2005.
[44]Wikipedia contributors. (2021). Monte Carlo localization. , 取自 https://en.wikipedia.org/w/index.php?title=Monte_Carlo_localization&oldid=997855519
[45]NickLamprianidis. (2018). costmap_2d., 取自 http://wiki.ros.org/costmap_2d
[46]ROBOTIS. (2021). DYNAMIXEL MX Series，取自 https://emanual.robotis.com/docs/en/dxl/
[47]YOLOv4 vs YOLOv4-tiny, Techzizou, https://medium.com/analytics-vidhya/yolov4-vs-yolov4-tiny-97932b6ec8ec , (accessed on 25,02,2021)
[48]2018，move_base，ros.org， Retrieved from: http://wiki.ros.org/move_base
[49]DOBOT CR5 Working Range, https://www.dobot-robots.com/products/cr-series/cr5.html