近年來工業 4.0 發展迅速，無人搬運車（Automated Guided Vehicle, AGV）的技術也越來越成熟，工廠無人化是未來的趨勢，本研究主旨為讓無人堆高機直接插取圓形貨品並進行堆疊。為了節省傳統堆高機運送貨品時將貨品固定上棧板的人工，並達到無人化，本研究主要分為四個程序:取貨程序、全域定位、牆面辨識、堆疊程序，為完成所有的程序，透過設計的前插機構使用IntelRealSense D435進行顏色與形狀的辨識。將從RealSense取得的深度值透過三角函數計算出圓形貨品與AGV的相對角度，將取得貨品傾斜角度進行AGV航向角的控制。全域定位則使用WebCam進行顏色辨識取得AGV位置後使用PID計算目標點，取得計算後的目標點帶入速度與角速度完成點對點的位移。堆疊時則使用LIDAR掃描將點使用向量的內積進行牆面的辨識取得AGV與牆面的相對角度與距離，並在AGV上安裝了FSR(Force-Sensitive Resistor)、近接開關、真空感測器，透過通訊進行資訊的傳遞可更加準確迅速的完成任務。
為了證明本論文方法之有效性，將實驗分成四個部分，第一個部分利用天花板WebCam使用PID進行多起點到終點，驗證定位精度與花費的時間，第二部分使用LIDAR透過2D掃描進行牆面的辨識，驗證固定距離多角度下牆面辨識的可靠度。第三個部分使用IntelRealSense深度相機進行取貨的極限角度與距離的測試，驗證取貨的有效性。第四個部分結合堆疊與取貨進行測試，驗證貨品堆疊與取貨的準確度與花費時間。

In recent years, Industry 4.0 has developed rapidly, and the technology of Automated Guided Vehicle (AGV) has become more and more mature. The unmanned factory is the future trend. In order to save the labor of fixing the goods on the pallet when the traditional stacker transports the goods as well as to achieve unmanned operation, this research is mainly divided into four procedures: the picking procedure, the global positioning, the wall identification, and the stacking procedure. The program uses Intel RealSense D435 for color and shapes recognition through the designed front insertion mechanism. The depth value obtained from RealSense is used to calculate the relative angle between the circular product and the AGV through the trigonometric function, and the inclination angle of the product is obtained to control the AGV heading angle. For global positioning, WebCam is used for color identification to obtain the position of the AGV, and then the PID is used to calculate the target point, and the calculated target point is brought into the velocity and angular velocity to complete the point-to-point displacement. When stacking, use LIDAR scanning to identify the wall using the inner product of the vector to obtain the relative angle and distance between the AGV and the wall, and install FSR (Force-Sensitive Resistor), proximity switches, and vacuum sensors on the AGV. the transmission of information through communication can complete the task more accurately and quickly.
In order to prove the effectiveness of the method in this thesis, the experiment is divided into four parts. The first part uses the ceiling WebCam to use PID to perform multiple start-to-end points to verify the positioning accuracy and time spent. The second part uses LIDAR to identify the wall through 2D scanning. to verify the reliability of wall identification under different- distances and angles. The third part uses the Intel RealSense depth camera to test the limit angle and distance of pickup to verify the effectiveness of pickup procedure. The fourth part combines stacking and picking to verify the accuracy and time spent on stacking and picking process.

[1] Z. Song, Y. Zhang, Y. Liu, K. Yang and M. Sun,"MSFYOLO: Feature fusion-based detection for small objects," in IEEE Latin America Transactions, vol. 20, no. 5, pp. 823-830, May 2022, DOI: 10.1109/TLA.2022.9693567.
[2] S. Li, Y. Li, Y. Li, M. Li and X. Xu, "YOLO-FIRI: Improved YOLOv5 for Infrared Image Object Detection," in IEEE Access, vol. 9, pp. 141861-141875, 2021, doi: 10.1109/ACCESS.2021.3120870.
[3] P. Shi, Z. Zhao, X. Fan, X. Yan, W. Yan and Y. Xin, "Remote Sensing Image Object Detection Based on Angle Classification," in IEEE Access, vol. 9, pp. 118696-118707, 2021, doi: 10.1109/ACCESS.2021.3107358.
[4] Y. Ai, T. Rui, M. Lu, L. Fu, S. Liu and S. Wang, "DDL-SLAM: A Robust RGB-D SLAM in Dynamic Environments Combined With Deep Learning," in IEEE Access, vol. 8, pp. 162335-162342, 2020, doi: 10.1109/ACCESS.2020.2991441.
[5] C. Arth, C. Pirchheim, J. Ventura, D. Schmalstieg and V. Lepetit, "Instant Outdoor Localization and SLAM Initialization from 2.5D Maps," in IEEE Transactions on Visualization and Computer Graphics, vol. 21, no. 11, pp. 1309-1318, 15 Nov. 2015, doi: 10.1109/TVCG.2015.2459772.
[6] D. Li et al., "A SLAM System Based on RGBD Image and Point-Line Feature," in IEEE Access, vol. 9, pp. 9012-9025, 2021, doi: 10.1109/ACCESS.2021.3049467.
[7] Y. Li, N. Brasch, Y. Wang, N. Navab and F. Tombari, "Structure-SLAM: Low-Drift Monocular SLAM in Indoor Environments," in IEEE Robotics and Automation Letters, vol. 5, no. 4, pp. 6583-6590, Oct. 2020, doi: 10.1109/LRA.2020.3015456.
[8] A. Fitzgibbon, M. Pilu and R. B. Fisher, "Direct least square fitting of ellipses," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, no. 5, pp. 476-480, May 1999, doi: 10.1109/34.765658.
[9] Y. Han, Y. Cheng and G. Xu, "Trajectory Tracking Control of AGV Based on Sliding Mode Control With the Improved Reaching Law," in IEEE Access, vol. 7, pp. 20748-20755, 2019, doi: 10.1109/ACCESS.2019.2897985.
[10] I. Draganjac, D. Miklić, Z. Kovačić, G. Vasiljević and S. Bogdan, "Decentralized Control of Multi-AGV Systems in Autonomous Warehousing Applications," in IEEE Transactions on Automation Science and Engineering, vol. 13, no. 4, pp. 1433-1447, Oct. 2016, doi: 10.1109/TASE.2016.2603781.
[11] P. R. Pinheiro et al., "Integration of the Mobile Robot and Internet of Things to Monitor Older People," in IEEE Access, vol. 8, pp. 138922-138933, 2020, doi: 10.1109/ACCESS.2020.3009167.
[12] M. Kokot, D. Miklić and T. Petrović, "Path Continuity for Multi-Wheeled AGVs," in IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 7437-7444, Oct. 2021, doi: 10.1109/LRA.2021.3099086.
[13] P. T. Lin, C. -A. Liao and S. -H. Liang, "Probabilistic Indoor Positioning and Navigation (PIPN) of Autonomous Ground Vehicle (AGV) Based on Wireless Measurements," in IEEE Access, vol. 9, pp. 25200-25207, 2021, doi: 10.1109/ACCESS.2021.3057415.
[14] D. Shi, H. Mi, E. G. Collins and J. Wu, "An Indoor Low-Cost and High-Accuracy Localization Approach for AGVs," in IEEE Access, vol. 8, pp. 50085-50090, 2020, doi: 10.1109/ACCESS.2020.2980364.
[15] J. Villagra and D. Herrero-Perez, "A Comparison of Control Techniques for Robust Docking Maneuvers of an AGV," in IEEE Transactions on Control Systems Technology, vol. 20, no. 4, pp. 1116-1123, July 2012, doi: 10.1109/TCST.2011.2159794.
[16] X. -X. Du, Y. Mu, Z. -W. Ye and Y. -J. Zhu, "A Passive Target Recognition Method Based on LED Lighting for Industrial Internet of Things," in IEEE Photonics Journal, vol. 13, no. 4, pp. 1-8, Aug. 2021, Art no. 7300308, doi: 10.1109/JPHOT.2021.3098672.
[17] T. Qiuyun, S. Hongyan, G. Hengwei and W. Ping, "Improved Particle Swarm Optimization Algorithm for AGV Path Planning," in IEEE Access, vol. 9, pp. 33522-33531, 2021, doi: 10.1109/ACCESS.2021.3061288.
[18] R. Chai, A. Tsourdos, A. Savvaris, S. Chai, Y. Xia and C. L. P. Chen, "Design and Implementation of Deep Neural Network-Based Control for Automatic Parking Maneuver Process," in IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 4, pp. 1400-1413, April 2022, doi: 10.1109/TNNLS.2020.3042120.
[19] D. Herrero, J. Villagrá and H. Martínez, "Self-Configuration of Waypoints for Docking Maneuvers of Flexible Automated Guided Vehicles," in IEEE Transactions on Automation Science and Engineering, vol. 10, no. 2, pp. 470-475, April 2013, doi: 10.1109/TASE.2013.2240386.
[20] Z. Rozsa and T. Sziranyi, "Obstacle Prediction for Automated Guided Vehicles Based on Point Clouds Measured by a Tilted LIDAR Sensor," in IEEE Transactions on Intelligent Transportation Systems, vol. 19, no. 8, pp. 2708-2720, Aug. 2018, doi: 10.1109/TITS.2018.2790264.
[21] J. S. Keek, S. L. Loh and S. H. Chong, "Comprehensive Development and Control of a Path-Trackable Mecanum-Wheeled Robot," in IEEE Access, vol. 7, pp. 18368-18381, 2019, doi: 10.1109/ACCESS.2019.2897013.
[22] L. Zamora-Cadenas, I. Velez and J. E. Sierra-Garcia, "UWB-Based Safety System for Autonomous Guided Vehicles Without Hardware on the Infrastructure," in IEEE Access, vol. 9, pp. 96430-96443, 2021, doi: 10.1109/ACCESS.2021.3094279.
[23] Q. Zou, Q. Sun, L. Chen, B. Nie and Q. Li, "A Comparative Analysis of LiDAR SLAM-Based Indoor Navigation for Autonomous Vehicles," in IEEE Transactions on Intelligent Transportation Systems, vol. 23, no. 7, pp. 6907-6921, July 2022, doi: 10.1109/TITS.2021.3063477.
[24] S. Lu, C. Xu and R. Y. Zhong, "An Active RFID Tag-Enabled Locating Approach With Multipath Effect Elimination in AGV," in IEEE Transactions on Automation Science and Engineering, vol. 13, no. 3, pp. 1333-1342, July 2016, doi: 10.1109/TASE.2016.2573595.
[25] Z. Deng, T. Zhang, D. Liu, X. Jing and Z. Li, "A High-Precision Collaborative Control Algorithm for Multi-Agent System Based on Enhanced Depth Image Fusion Positioning," in IEEE Access, vol. 8, pp. 34842-34853, 2020, doi: 10.1109/ACCESS.2020.2973344.
[26] J. H. Cho and M. -W. Cho, "Effective Position Tracking Using B-Spline Surface Equation Based on Wireless Sensor Networks and Passive UHF-RFID," in IEEE Transactions on Instrumentation and Measurement, vol. 62, no. 9, pp. 2456-2464, Sept. 2013, doi: 10.1109/TIM.2013.2259099.
[27] R. Szczepanski, T. Tarczewski and K. Erwinski, "Energy Efficient Local Path Planning Algorithm Based on Predictive Artificial Potential Field," in IEEE Access, vol. 10, pp. 39729-39742, 2022, doi: 10.1109/ACCESS.2022.3166632.
[28] S. Thakar, P. Rajendran, A. M. Kabir and S. K. Gupta, "Manipulator Motion Planning for Part Pickup and Transport Operations From a Moving Base," in IEEE Transactions on Automation Science and Engineering, vol. 19, no. 1, pp. 191-206, Jan. 2022, doi: 10.1109/TASE.2020.3020050.
[29] W. Li and R. Xiong, "Dynamical Obstacle Avoidance of Task- Constrained Mobile Manipulation Using Model Predictive Control," in IEEE Access, vol. 7, pp. 88301-88311, 2019, doi: 10.1109/ACCESS.2019.2925428.
[30] B. P. L. Lau et al., "Multi-AGV's Temporal Memory-Based RRT Exploration in Unknown Environment," in IEEE Robotics and Automation Letters, vol. 7, no. 4, pp. 9256-9263, Oct. 2022, doi: 10.1109/LRA.2022.3190628.