本研究為使用點雲形式的資料來進行掃描數據的處理與3D顯示，再透過隨機森林學習演算法、KNN學習演算法和支持向量機來判斷在圓管上的焊道，提取少量資訊作為機械手臂模擬軟體RobotMaster的輸入資料來進行機械手臂在圓管上焊道研磨路徑的規劃與生成。在機械工程領域中焊道研磨為一項較耗時但又不可缺的加工過程，由於對加工品質的要求及研磨時所產生的微粒容易對加工者的健康產生影響等因素，使近期用機械手臂代替真人進行研磨工作的方式越來越多。由於在製造上存在各種不穩定的狀況，即便相同的物件其幾何形狀(包含長度、形狀等)也會有些許差異，導致機械手臂在進行相同物件的加工時無法使用相同路徑進行加工，這間接造成加工路徑規劃的成本上升。在以往的研究中使用許多不同類型的感測器來分析待加工的工件，其中包含深度感測器、紅外線掃描器及各種二維圖像等，但大多感測器都存在著外界環境造成的影響與精度上的問題。近期也開始使用點雲資料來分析待加工的工件，然而對於點雲資料的處理大多需要消耗大量的計算資源與時間成本。為了提高效率，本研究提出了一種點雲資料前處理、焊道辨識以及機械手臂加工路徑規劃的系統並結合機械手臂模擬軟體RobotMaster進行碰撞模擬及加工路徑生成，在不進行工件與焊道3D輪廓重建的狀況下，以相對於以往來說成本較低的方法來完成目標。研究結果顯示在平面與曲面上的焊道辨識中使用隨機森林學習演算法辨識的準確率能維持在98.5%以上，使用KNN學習演算法辨識的準確率能維持在97%以上，使用支持向量機辨識的準確率能維持在96.5%以上。實際研磨驗證結果與期望結果誤差為0.0660mm。

The aim of this study is to identify the geometry of weld beads from laser scanner data for robotic grinding planning. Robotic grinding is used widely in the industry be-cause of the dirty environment and healthy issue. However, the some of the grinding op-eration re-quires intelligent recognition of the working area because of the varying geom-etry of the workpieces, such as weld beads. Previous studies have used various sensors to analyze workpieces, including depth sensors, infrared scanners, and 2D images. Howev-er, laser line scanner has the highest accuracy among them. In order to extract the geome-try and position of weld beads from the point data, three classification algorithms, Ran-dom Forest, k-nearest neighborhood and Supported Vector Machine, were studied and compared. The results show that the accuracy of identifying weld beads using the random forest learning algorithm is 98.98%, using the KNN learning algorithm is 97.34%, and using the support vector machine is 97.02%. All the three models provided the maximum height standard deviation of 0.01293 mm. The RF measured the average width standard deviation of 0.08991 mm. The average width standard deviation of 0.07772 mm was ob-tained by KNN. SVM detected the average width standard deviation of 0.07723 mm. An excellent grinding error of 0.066 mm was achieved.

[1] J. Li, "Application of Industrial Robots in Grinding Technology through Path Plan-ning and Artificial Intelligence," 2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT), Chongqing, China, 22-24 November 2021, pp. 851-854, doi: 10.1109/ICESIT53460.2021.9696769.
[2] P. Lv et al., "Design of Intelligent Watering Robot," 2020 IEEE International Confer-ence on Mechatronics and Automation (ICMA), Beijing, China, 13-16 October 2020, pp. 1499-1504, doi: 10.1109/ICMA49215.2020.9233656.
[3] Y. Du, Z. Z. Yuan, X. Tian and Z. Yang, "Clothing gripping and sorting based on a dual-arm collaborative robot," 2023 International Conference on Robotics and Auto-mation in Industry (ICRAI), Peshawar, Pakistan, 03-05 March 2023, pp. 1-6, doi: 10.1109/ICRAI57502.2023.10089595.
[4] S. Wang, R. Wang, P. Bai and X. Wang, "Modeling and Data Analysis of Robotic Arm-Assisted Device for Detection of Violently Infectious Viruses," 2023 IEEE 11th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), Chongqing, China, 08-10 December 2023, pp. 1413-1418, doi: 10.1109/ITAIC58329.2023.10409038.
[5] M. Jinno et al., "Development of a force controlled robot for grinding, chamfering and polishing," Proceedings of 1995 IEEE International Conference on Robotics and Au-tomation, Nagoya, Japan, 21-27 May 1995, pp. 1455-1460 vol.2, doi: 10.1109/ROBOT.1995.525481.
[6] F. Cao, Y. Li, G. Zhang, J. Wang, X. Chen and Y. Zhao, "Novel humanoid dual-arm grinding robot," 2016 12th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), Auckland, New Zealand, 29-31 August 2016, pp. 1-6, doi: 10.1109/MESA.2016.7587185.
[7] K. -H. Liu and M. -F. Chen, "Research and development of robot arm applied to grinding path planning of metal parts," 2022 IET International Conference on Engi-neering Technologies and Applications (IET-ICETA), Changhua, Taiwan, 14-16 Oc-tober 2022, pp. 1-2, doi: 10.1109/IET-ICETA56553.2022.9971597.
[8] J. Peng, J. Han and Y. Yang, "Research on Mechanical Arm Grinding Method Based on Improved CascadePSP Net," 2022 7th International Conference on Communica-tion, Image and Signal Processing (CCISP), Chengdu, China, 18-20 November 2022, pp. 33-38, doi: 10.1109/CCISP55629.2022.9974445.
[9] Y. T. Wang and Y. J. Jan, "Path planning for robot-assisted grinding processes," Pro-ceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164), Seoul, Korea (South), 21-26 May 2001, vol.1, pp. 331-336, doi: 10.1109/ROBOT.2001.932573.
[10] K. -H. Liu and M. -F. Chen, "Research and development of robot arm applied to grinding path planning of metal parts," 2022 IET International Conference on Engi-neering Technologies and Applications (IET-ICETA), Changhua, Taiwan, 14-16 Oc-tober 2022, pp. 1-2, doi: 10.1109/IET-ICETA56553.2022.9971597.
[11] X. Wang, D. Liu, Y. Tao and Y. Cui, "An Optimized Path Planning Method for Off-Line Programming of a Industrial Robot," 2012 International Conference on Computer Science and Electronics Engineering, Hangzhou, China, 23-25 March 2012, pp. 57-60, doi: 10.1109/ICCSEE.2012.159.
[12] Z. Wang, N. Zhang and A. Wang, "Research on Trajectory Planning Algorithm for Intelligence Robot Cutting-Grinding," 2023 IEEE International Conference on Mech-atronics and Automation (ICMA), Harbin, Heilongjiang, China, 06-08 August 2023, pp. 233-238, doi: 10.1109/ICMA57826.2023.10215946.
[13] G. Zhang, J. Wang, F. Cao, Y. Li and X. Chen, "3D curvature grinding path planning based on point cloud data," 2016 12th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), Auckland, New Zealand, 29-31 August 2016, pp. 1-6, doi: 10.1109/MESA.2016.7587150.
[14] Yang Zhao, Ji Zhao, Lei Zhang, Lizhe Qi and Qing Tang, "Path planning for auto-matic robotic blade grinding," 2009 International Conference on Mechatronics and Automation, Changchun, 09-12 August 2009, pp. 1556-1560, doi: 10.1109/ICMA.2009.5246534.
[15] S. Diao, X. Chen and L. Wu, "Task-Level Time-Optimal Machining Path Planning for Grinding Manipulators," 2018 IEEE 9th International Conference on Software Engineering and Service Science (ICSESS), Beijing, China, 23-25 November 2018, pp. 825-830, doi: 10.1109/ICSESS.2018.8663821.
[16] Y. Xie, J. Yang, M. Feng, W. Huang and J. Li, "Path planning of grinding robot with force control based on B-spline curve," 2019 IEEE International Conference on Ro-botics and Biomimetics (ROBIO), Dali, China, 06-08 December 2019, pp. 2732-2736, doi: 10.1109/ROBIO49542.2019.8961568.
[17] N. Wang, Q. Wang, Q. Zhang and J. Xie, "Adaptive Grinding Planning of Robotic Arms With Minimal Cost," in IEEE Transactions on Instrumentation and Measure-ment, vol. 73, pp. 1-16, 2024, Art no. 2508016, doi: 10.1109/TIM.2024.3364268.
[18] Q. Liu, W. Sun, H. Yuan, N. Liu and S. Shu, "Weld Point Cloud Segmentation Algo-rithm Fused Contour Feature," 2021 China Automation Congress (CAC), Beijing, China, 22-24 October 2021, pp. 1185-1188, doi: 10.1109/CAC53003.2021.9728533.
[19] Y. Liu and P. Sun, "Deep Learning Point Cloud Classification Algorithm Considering Local Spatial Features," 2023 4th International Conference on Computer Engineering and Application (ICCEA), Hangzhou, China, 07-09 April 2023, pp. 835-838, doi: 10.1109/ICCEA58433.2023.10135361.
[20] W. Zhang, Y. Yu and F. Yang, "A Novel Grid-Based Geometry Compression Frame-work for Spinning Lidar Point Clouds," 2022 IEEE International Conference on Mul-timedia and Expo (ICME), Taipei, Taiwan, 18-22 July 2022, pp. 1-6, doi: 10.1109/ICME52920.2022.9858927.
[21] J. Yu, F. Liu, Y. Li, Y. Shi, X. Gao and K. Zhao, "Research on ICESat-2 Single Pho-ton Laser Point Cloud Denoising and Classification Algorithm," 2023 5th Internation-al Conference on Geoscience and Remote Sensing Mapping (GRSM), Lianyungang, China, 13-15 October 2023, pp. 40-45, doi: 10.1109/GRSM60169.2023.10425042.
[22] J. Yao, C. Qian, G. Yu and Y. Zhang, "A 3D Reconstruction Method for Sheet Parts Using Binocular Camera," 2023 IEEE International Conference on Mechatronics and Automation (ICMA), Harbin, Heilongjiang, China, 06-09 August 2023, pp. 663-668, doi: 10.1109/ICMA57826.2023.10216024.
[23] Q. -L. Cai, K. -W. Chen, C. -Y. Yao and H. -K. Chu, "Automatic Local Point Cloud Registration Algorithm and Point Cloud Reconstruction System," 2021 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS), Hualien City, Taiwan, 16-19 November 2021, pp. 1-2, doi: 10.1109/ISPACS51563.2021.9651113.
[24] H. Zeng, Y. Chen, Z. Zhang, C. Wang and J. Li, "Reconstruction of 3D Zebra Cross-ings from Mobile Laser Scanning Point Clouds," IGARSS 2019 - 2019 IEEE Interna-tional Geoscience and Remote Sensing Symposium, Yokohama, Japan, 28 July 2019 – 02 August 2019, pp. 1899-1902, doi: 10.1109/IGARSS.2019.8899336.
[25] W. Wu, L. Kong, W. Liu and C. Zhang, "Laser Sensor Weld Beads Recognition and Reconstruction for Rail Weld Beads Grinding Robot," 2017 5th International Con-ference on Mechanical, Automotive and Materials Engineering (CMAME), Guang-zhou, China, 01-03 August 2017, pp. 143-148, doi: 10.1109/CMAME.2017.8540113.
[26] Y. Li, Y. F. Li, Q. L. Wang, D. Xu and M. Tan, "Measurement and Defect Detection of the Weld Bead Based on Online Vision Inspection," in IEEE Transactions on In-strumentation and Measurement, vol. 59, no. 7, pp. 1841-1849, July 2010, doi: 10.1109/TIM.2009.2028222.
[27] Guoliang Ye, Jianwen Guo, Zhenzhong Sun, Chuan Li, Shouyan Zhong, Weld bead recognition using laser vision with model-based classification, Robotics and Comput-er-Integrated Manufacturing, Volume 52, 2018, Pages 9-16, ISSN 0736-5845, https://doi.org/10.1016/j.rcim.2018.01.006.
[28] Feng, H., Ren, X., Li, L. et al. A novel feature-guided trajectory generation method based on point cloud for robotic grinding of freeform welds. Int J Adv Manuf Tech-nol 115, 1763–1781 (2021). https://doi.org/10.1007/s00170-021-07095-2
[29] Ge, J., Deng, Z., Li, Z. et al. An efficient system based on model segmentation for weld seam grinding robot. Int J Adv Manuf Technol 121, 7627–7641 (2022). https://doi.org/10.1007/s00170-022-09758-0
[30] Ge, J., Deng, Z., Wang, S. et al. Vision Sensing-Based Online Correction System for Robotic Weld Grinding. Chin. J. Mech. Eng. 36, 125 (2023). https://doi.org/10.1186/s10033-023-00955-w
[31] M. D. Kamboj, M. Singh, M. Usmani, S. Ahmad and M. A. Gaur, "A Comparative Analysis of Algorithms for Effective Waste Classification," 2023 International Con-ference on Sustainable Communication Networks and Application (ICSCNA), Theni, India, 15-17 November 2023, pp. 598-604, doi: 10.1109/ICSCNA58489.2023.10370654.
[32] B. Zhong, M. S. Sainin and T. S. Fun, "Knowledge Base Processing Method Based on Text Classification Algorithm," 2023 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET), Kota Kinabalu, Malaysia, 12-14 September 2023, pp. 96-100, doi: 10.1109/IICAIET59451.2023.10291339.
[33] A. N. Islami, A. Prasetiadi and N. G. Ramadhan, "Hemorrhagic Stroke Classification on Computerized Tomography Scan Images of the Brain using CNN Algorithm," 2023 IEEE International Conference on Communication, Networks and Satellite (COMNETSAT), Malang, Indonesia, 23-25 November 2023, pp. 239-244, doi: 10.1109/COMNETSAT59769.2023.10420560.
[34] D. Velusamy, A. Mounika, D. Ragulraj and K. Vigneshwaran, "Automatic Disease Classification of Paddy Leaf Images Using Deep Learning Algorithms," 2023 4th In-ternational Conference on Signal Processing and Communication (ICSPC), Coimba-tore, India, 23-24 March 2023, pp. 325-329, doi: 10.1109/ICSPC57692.2023.10125872.
[35] Breiman, L. (2001) Random Forests. Machine Learning, 45, 5-32. https://doi.org/10.1023/A:1010933404324.
[36] S. U. Raj, M. Veera Manikanta, P. S. Sai Harsitha and M. Judith Leo, "Vacant Parking Lot Detection System Using Random Forest Classification," 2019 3rd International Conference on Computing Methodologies and Communication (ICCMC), Erode, In-dia, 27-29 March 2019, pp. 454-458, doi: 10.1109/ICCMC.2019.8819689.
[37] Fix, Evelyn, and J. L. Hodges. “Discriminatory Analysis. Nonparametric Discrimina-tion: Consistency Properties.” International Statistical Review / Revue Internationale de Statistique, vol. 57, no. 3, 1989, pp. 238–47. JSTOR, https://doi.org/10.2307/1403797.
[38] Cortes, C., Vapnik, V. Support-vector networks. Mach Learn 20, 273–297 (1995). https://doi.org/10.1007/BF00994018.
[39] C. Zhang, X. Sun, J. Hu and W. Deng, "Precise eye localization by fast local linear SVM," 2014 IEEE International Conference on Multimedia and Expo (ICME), Chengdu, China, 14-18 September 2014, pp. 1-6, doi: 10.1109/ICME.2014.6890174.
[40] 郭家維。「使用KNN為機械手臂研磨製程從點雲資料中進行焊道判斷」。碩士論文，國立成功大學智慧製造國際碩士學位學程，2023。https://hdl.handle.net/11296/9x7b2t。