臺灣博碩士論文加值系統

菇類栽培環境中的溫度與濕度、二氧化碳濃度及病蟲害是栽培作業中需要控制的重要參數。傳統巡檢作業需要頻繁地到栽培場域進行實地觀察與記錄分析，既耗費人力也要求人員須有專業領域知識。本研究嘗試設計一自主移動機器人(AMR)平台，使其能自主於菇房中定點巡檢、擷取資料及污染辨識，並可在切換手動模式後，快速移動至指定位置以便於後續遠端作業之執行。本AMR平台以機器人作業系統(Robot Operating System, ROS)進行設計與整合，並安裝於樹莓派4版之單板電腦上，平台中置一組垂直定位系統，裝設上熱顯像與相機等感測模組，使之能沿栽培層架高度方向移動。本研究採用YDLidar及MPU6050作為地圖建構之感測器，並以Cartographer演算法進行同步定位與地圖建構(Simultaneous localization and mapping, SLAM)。為了獲得最佳的位置資訊，本移動平台在首次到達定位後，系統會掃描附近直到尋獲指定Apriltag圖碼，並在計算與其之距離誤差後再次移動，做第二次定位。本研究利用YOLO物件偵測模型，可對菇類太空包進行污染辨識，在最終成果中，其預測精準度可達90 %以上。本研究利用遠端postgresql資料庫對巡檢過程中取得的所有資料進行資料庫新增、讀取及刪除等操作，便於後續大數據分析與巡檢實驗時資料的取用。本研究之自主移動機器人平台可成功的在實作場域中操作，對解決菇類產業巡檢作業自動化問題有重要的貢獻。

Temperature, humidity, concentration of carbon dioxide and pests and diseases are important parameters in the mushroom cultivation environment. Early on, inspection require workers frequent visit the cultivation field for record and analysis. That consumes manpower and requires workers to have professional knowledge. This dissertation attempts to design an autonomous mobile robot (AMR) platform and require it able to autonomously conduct fixed-point inspection in the mushroom house, data acquisition, pollution detect, and quick move to the specify location after switching the manual mode for subsequent remote operations. This mobile platform is designed and integrated with the Robot Operating System (ROS), and installed on the Raspberry Pi 4. A set of vertical positioning system with sensors is installed in the platform. In this dissertation, simultaneous localization and map construction (Simultaneous localization and mapping, SLAM) use YDLidar and MPU6050 as sensor and cartographer as algorithm. In order to obtain the best location information, after the mobile platform arrives at the location for the first time, the system will scan until the specified Apriltag code is found, then move again for the second location. YOLO is used for dectet the pollution of mushroom bag in this dissertation.The test result, the precision is better than 90%. All data acquired during the inspection will be created, read and deleted in remote postgresql database. In this way, data can be accessed more conveniently in subsequent operations. This autonomous mobile robot platform has been successfully operated in the mushroom house. For benefits mushroom industry to solving the automation problem of inspection.

摘要 ……………………………………………………………………………… i
Abstract …………………………………………………………………………. ii
誌謝 ……………………………………………………………………………… iii
目錄 ………………………………………………………………………………. iv
表目錄 ……………………………………………………………………………. vi
圖目錄 …………………………………………………………………………… vii
第一章 緒論 ………………………………………………………………………. 1
1.1 前言 …………………………………………………………………………… 1
1.2 研究背景 ……………………………………………………………………… 1
1.3 研究動機與目的 ……………………………………………………………… 2
1.4 論文架構 ……………………………………………………………………… 2
第二章 文獻回顧 …………………………………………………………………. 3
2.1 袋栽菇類完整生產流程 ……………………………………………………… 3
2.2 農業自動化 …………………………………………………………………… 4
2.2.1 環境監控 …………………………………………………………………… 4
2.2.2 農業機器人 ………………………………………………………………… 5
2.2.3 農業人機協作(Human–robot interaction in agriculture) ………………. 7
2.3 機器人作業系統 (Robot Operating System, ROS) ………………………. 8
2.3.1 同時定位與建圖(Simultaneous Localization And Mapping, SLAM) … 10
2.3.2 路徑規劃器 …………………………………………………………………. 12
2.4 電腦視覺於菇類栽培 …………………………………………………………. 13
2.4.1 袋栽菇類於生長過程中之異常狀況 ………………………………………. 13
2.4.2 YOLO ………………………………………………………………………. 13
第三章 研究方法 …………………………………………………………………. 15
3.1 巡檢規劃 ………………………………………………………………………. 15
3.2 AMR系統架構 ………………………………………………………………… 16
3.3 控制系統 ………………………………………………………………………. 17
3.3.1 邊緣運算控制系統 …………………………………………………………. 17
3.3.2 上層控制系統 ………………………………………………………………. 17
3.3.3 下層控制系統 ………………………………………………………………. 18
3.4 移動系統 ………………………………………………………………………. 19
3.4.1 移動模型 ……………………………………………………………………. 19
3.4.2 硬體設計 ……………………………………………………………………. 20
3.4.3 ROS程式控制 ……………………………………………………………… 21
3.4.4 里程計(Odometry) …………………………………………………………. 22
3.5 層架垂直定位系統 ……………………………………………………………. 24
3.5.1 硬體設計 ……………………………………………………………………. 24
3.5.2 ROS程式控制 ……………………………………………………………… 25
3.6 環境感測系統 …………………………………………………………………. 26
3.6.1 影像擷取單元 ………………………………………………………………. 26
3.6.2 氣候擷取單元 ………………………………………………………………. 28
3.7 導航系統 ………………………………………………………………………. 29
3.7.1 AMR移動參數設置 ………………………………………………………… 29
3.7.2 光學雷達(Light Detection And Ranging, LiDAR)設置 …………………. 30
3.7.3 Cartographer設置 …………………………………………………………. 31
3.7.4 Navigation Stack ………………………………………………………….. 32
3.7.5 多點路徑規劃 ………………………………………………………………. 35
3.7.6 二次定位 ……………………………………………………………………. 37
3.8 Postgresql資料庫處理 ………………………………………………………. 38
3.8.1 資料新增 ……………………………………………………………………. 38
3.8.2 資料刪除 ……………………………………………………………………. 39
3.8.3 資料讀取 ……………………………………………………………………. 40
3.9 汙染辨識 ………………………………………………………………………. 41
3.9.1 訓練環境設置 ………………………………………………………………. 41
3.9.2 Scaled-YOLOv4訓練設置 ………………………………………………… 42
3.9.3 YOLOv7訓練設置 …………………………………………………………. 43
3.10 UI介面設計 …………………………………………………………………. 44
3.10.1 ros2djs介紹 ………………………………………………………………. 44
3.10.2 Flask介紹 …………………………………………………………………. 46
第四章 結果與討論 ………………………………………………………………. 48
4.1 AMR 性能測試 ……………………………………………………………….. 50
4.2 實際菇場地圖建置 ……………………………………………………………. 51
4.3 巡檢功能 ………………………………………………………………………. 52
4.3.1 導航控制 ……………………………………………………………………. 52
4.3.2 二次定位測試 ………………………………………………………………. 53
4.3.3 多點路徑規劃 ………………………………………………………………. 57
4.4 環境資料擷取 …………………………………………………………………. 58
4.5 菇包汙染辨識 …………………………………………………………………. 59
4.5.1 Scaled-YOLOv4辨識結果 ………………………………………………… 59
4.5.2 YOLOv7辨識結果 …………………………………………………………. 60
4.5.3 小結 …………………………………………………………………………. 60
第五章 結論與未來展望 …………………………………………………………… 61
5.1 結論 ……………………………………………………………………………. 61
5.2 未來展望 ………………………………………………………………………. 61
參考文獻 ……………………………………………………………………………. 62
Extended Abstract ………………………………………………………………… 65

[1]方怡丹、蔡清榮，2016，“臺灣菇類產業之發展與輔導”，菇類生技產業研討會專刊。
[2]黃燿聖，2013，“菇類太空包自動化套袋模組之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[3]陳柏翰，2013，“液態接種噴嘴系統之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[4]阮忠興，2015，“創新菇類智慧製造產線之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[5]李志仁，2017，“創新袋式液態菌種接菌系統之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[6]蔡孟佑，2018，“菇類太空包自動化生產作業模擬”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[7]劉瑋珊，2015，“太空包束口翻袋機構之創新設計”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[8]黃文信，2018，“基於深度學習之杏鮑菇自動化分級系統之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[9]蘇爾曼，2019，“創新菇類栽培物聯網”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[10]林明毅，2019，“杏鮑菇自動化採收機構之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[11]古丙駿，2019，“杏鮑菇採收深度學習物件辨識系統之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[12]顏靖峰，2021，“袋栽杏鮑菇完株採收機構之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[13]胡原愷，2020，“袋栽菇類製包流程智慧檢測系統之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[14]邱弘丞，2020，“自主移動式智慧物聯模組在菇類栽培之研究”，國立虎尾科技大學機械設計工程系碩士班碩士論文，雲林縣。
[15]周榮源、石信德、黃文信，2020，“袋栽菇類製包技術回顧與智慧化發展初探”，行政院農委會農業科技決策資訊平台。2022年10月，取自https://agritech-foresight.atri.org.tw/article/contents/3144
[16]Auernhammer, Hermann, 2001, "Precision farming—the environmental challenge", Computers and electronics in agriculture, 30.1-3, pp. 31-43.
[17]Moysiadis, Vasileios, et al., 2021, "Smart farming in Europe", Computer science review, 39, 100345.
[18]Cambra Baseca, Carlos, et al., 2019, "A smart decision system for digital farming", Agronomy, 9, no. 5: 216.
[19]De Clercq, Matthieu, Anshu Vats, and Alvaro Biel, 2018, "Agriculture 4.0: The future of farming technology", Proceedings of the World Government Summit, pp. 11-13.
[20]K. Rawal, G. Gabrani, 2020, "Iot based computing to monitor indoor plants by using smart pot", SSRN 3562964.
[21]M. J. C. Samonte, E. P. E. Signo, et al., 2019, "Phyto: An iot urban gardening mobile app", Proceedings of the 2019 2nd International Conference on Information Science and Systems, pp. 135-139.
[22]N. Abd Rahim, F. A. Zaki, et al., 2020, "Smart app for gardening monitoring system using iot technology", system, vol. 29, no. 04, pp. 7375-7384.
[23]Owen Bawden, et al., 2017, "Robot for weed species plant-specific management", Special Issue: Agricultural Robotics, pp. 1179-1199.
[24]Christopher Lehnert, et al., 2017, "Autonomous Sweet Pepper Harvesting for Protected Cropping Systems", IEEE Robotics and Automation Letters, vol. 2, no. 2, pp. 872-879.
[25]Nguyen Duc Tai, Nguyen Truong Thinh, 2019, "Design of Cable Measuring System of a Robot Spraying Pesticides in Agricultural Farm", 2019 International Conference on System Science and Engineering, pp. 577-580.
[26]Xunchen Liu, et al., 2022, "Intelligent UAV platform: assist construction of agricultural production automation", 2022 7th International Conference on Intelligent Computing and Signal Processing (ICSP), pp. 1009-1012.
[27]Henry A.M. Williams, et al., 2019, "Robotic kiwifruit harvesting using machine vision, convolutional neural networks and robotic arms", Biosystems Engineering, Volume 181, pp. 140-156.
[28]Georgi Komitov, et al., 2022, "About the methodology for working a robot to destroy weeds", 2022 8th International Conference on Energy Efficiency and Agricultural Engineering, pp. 1-4.
[29]Vasconez, et al., 2019, "Human–robot interaction in agriculture: A survey and current challenges", Biosystems engineering 179, pp. 35-48.
[30]Adamides, Georgios, 2016, "User interfaces for human-robot interaction: Application on a semi-autonomous agricultural robot sprayer", PhD Doctoral Dissertation, Open University of Cyprus.
[31]R. C. Luo, et al., 2020, "Modular ROS Based Autonomous Mobile Industrial Robot System for Automated Intelligent Manufacturing Applications", 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM), pp. 1673-1678.
[32]L. Arcos, et al., 2020, "ROS based Experimental Testbed for Multi-Robot Formation Control", 2020 IEEE ANDESCON, pp. 1-6.
[33]A. Belzunce, et al., 2016, "Control system design of a teleoperated omnidirectional mobile robot using ROS", 2016 IEEE 11th Conference on Industrial Electronics and Applications (ICIEA), pp. 1283-1287.
[34]G. Anand, et al., 2017, "A sensor framework for human-robot collaboration in industrial robot work-cell", 2017 International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT), pp. 715-720.
[35]R. A. Atmoko and D. Yang, 2018, "Online Monitoring & Controlling Industrial Arm Robot Using MQTT Protocol", 2018 IEEE International Conference on Robotics, Biomimetics, and Intelligent Computational Systems (Robionetics ), pp. 12-16.
[36]Z. Al-Mashhadani, et al., 2020, "Autonomous Exploring Map and Navigation for an Agricultural Robot", 2020 3rd International Conference on Control and Robots (ICCR), pp. 73-78.
[37]V. R. Ponnambalam, et al., 2020, "Agri-Cost-Maps - Integration of Environmental Constraints into Navigation Systems for Agricultural Robots", 2020 6th International Conference on Control, Automation and Robotics (ICCAR), pp. 214-220.
[38]S. Kalaivanan and R. Kalpana, 2017, "Coverage path planning for an autonomous robot specific to agricultural operations", 2017 International Conference on Intelligent Computing and Control (I2C2), pp. 1-5.
[39]W. Hess, D. Kohler, et al., 2016, "Real-time loop closure in 2D LIDAR SLAM," 2016 IEEE International Conference on Robotics and Automation (ICRA), pp. 1271-1278.
[40]Qu, Pengtao, et al., 2021, "Mapping performance comparison of 2D SLAM algorithms based on different sensor combinations", Journal of Physics: Conference Series, Vol. 2024, No. 1, p. 012056.
[41]行政院農業委員會農業知識入口網，2017年8月，杏鮑菇栽培技術（二）。2022年12月，取自https://kmweb.coa.gov.tw/theme_data.php?theme=news&sub_theme=variety&id=54786
[42]Redmon, Joseph, et al., 2016, "You only look once: Unified, real-time object detection", Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 779-788.
[43]YOLO, 2016, YOLO: Real-Time Object Detection. Retrieved May, 2022, from https://pjreddie.com/darknet/yolov2/
[44]Z. Tang, et al., 2021, "Pest-YOLO: Deep Image Mining and Multi-Feature Fusion for Real-Time Agriculture Pest Detection", 2021 IEEE International Conference on Data Mining (ICDM), Auckland, pp. 1348-1353.
[45]M. Lippi, et al., 2021, "A YOLO-Based Pest Detection System for Precision Agriculture", 2021 29th Mediterranean Conference on Control and Automation (MED), PUGLIA, pp. 342-347.
[46]V. Ponnusamy, et al., 2020, "Smart Glass: Real-Time Leaf Disease Detection using YOLO Transfer Learning", 2020 International Conference on Communication and Signal Processing (ICCSP), pp. 1150-1154.
[47]C. Song, et al., 2020, "Automatic Detection and Image Recognition of Precision Agriculture for Citrus Diseases", 2020 IEEE Eurasia Conference on IOT, Communication and Engineering (ECICE), pp. 187-190.
[48]Wang Chien-Yao, Alexey Bochkovskiy, Hong-Yuan Mark Liao, 2021, "Scaled-yolov4: Scaling cross stage partial network", Proceedings of the IEEE/cvf conference on computer vision and pattern recognition, pp. 13029-13038.
[49]Wang Chien-Yao, Alexey Bochkovskiy, Hong-Yuan Mark Liao, 2022, "YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors", arXiv preprint arXiv:2207.02696.
[50]Wang Chien-Yao, 2022, "yolov7". Retrieved October, 2022, from https://github.com/WongKinYiu/yolov7
[51]ODrive Robotics, 2022, "ODrive Pro Documentation". Retrieved December, 2020, from https://docs.odriverobotics.com/v/latest/getting-started.html
[52]Malu, et al., 2014, "Kinematics, localization and control of differential drive mobile robot", Global Journals of Research in Engineering, 14(H1), pp. 1-7.
[53]YDLIDAR, 2020 , "ydlidar_ros_driver". Retrieved October, 2021, from https://github.com/YDLIDAR/ydlidar_ros_driver
[54]Swarn Avinash Kumar, Wolfgang Hess, 2022 , "Cartographer ROS". Retrieved December, 2021, from https://google-cartographer-ros.readthedocs.io/en/latest/compilation.html
[55]ROS.org, 2020, "Navigation". Retrieved December, 2020, from http://wiki.ros.org/navigation
[56]Wang Chien-Yao, 2021, "ScaledYOLOv4". Retrieved December, 2021, from https://github.com/WongKinYiu/ScaledYOLOv4/archive/yolov4-csp