請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/94063完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 吳日騰 | zh_TW |
| dc.contributor.advisor | Rih-Teng Wu | en |
| dc.contributor.author | 陳泓瑋 | zh_TW |
| dc.contributor.author | Hung-Wei Chen | en |
| dc.date.accessioned | 2024-08-14T16:30:22Z | - |
| dc.date.available | 2024-08-15 | - |
| dc.date.copyright | 2024-08-13 | - |
| dc.date.issued | 2024 | - |
| dc.date.submitted | 2024-08-10 | - |
| dc.identifier.citation | [1] Jianghai Liao, Yuanhao Yue, Dejin Zhang, Wei Tu, Rui Cao, Qin Zou, and Qingquan Li. Automatic tunnel crack inspection using an efficient mobile imaging module and a lightweight cnn. IEEE Transactions on Intelligent Transportation Systems, 23(9):15190–15203, 2022.
[2] Fu-Chen Chen and Mohammad R. Jahanshahi. Nb-cnn: Deep learning-based crack detection using convolutional neural network and naïve bayes data fusion. IEEE Transactions on Industrial Electronics, 65(5):4392–4400, 2018. [3] Sayyed Bashar Ali, Reshul Wate, Sameer Kujur, Anurag Singh, and Santosh Kumar. Wall crack detection using transfer learning-based cnn models. In 2020 IEEE 17th India Council International Conference (INDICON), pages 1–7, 2020. [4] Sara Yasmine Ouerk, Olivier Vo Van, and Mouadh Yagoubi. Rail crack propaga- tion forecasting using multi-horizons rnns. In Georgiana Ifrim, Romain Tavenard, Anthony Bagnall, Patrick Schaefer, Simon Malinowski, Thomas Guyet, and Vin- cent Lemaire, editors, Advanced Analytics and Learning on Temporal Data, pages 260–275, Cham, 2023. Springer Nature Switzerland. [5] Divya Gupta, Gaurav Goel, Navdeep Kaur, and Dapinder Kaur. Efficient concrete crack detection system using surf and rnn algorithm. 2021.97 doi:10.6342/NTU202403413 [6] Eftychios Protopapadakis, Athanasios Voulodimos, Anastasios Doulamis, Nikolaos Doulamis, and Tania Stathaki. Automatic crack detection for tunnel inspection using deep learning and heuristic image post-processing. Applied intelligence, 49:2793–2806, 2019. [7] Sung-suk Choi and Eung-kon Kim. Building crack inspection using small uav.In 2015 17th International Conference on Advanced Communication Technology(ICACT), pages 235–238, 2015. [8] Yanxiang Li, Jinming Ma, Ziyu Zhao, and Gang Shi. A novel approach for uav image crack detection. Sensors, 22(9), 2022. [9] Bin Lei, Ning Wang, Pengcheng Xu, and Gangbing Song. New crack detection method for bridge inspection using uav incorporating image processing. Journal of Aerospace Engineering, 31(5):04018058, 2018. [10] Yonas Zewdu Ayele, Mostafa Aliyari, David Griffiths, and Enrique Lopez Droguett. Automatic crack segmentation for uav-assisted bridge inspection. Energies, 13(23),2020. [11] Xiong Peng, Xingu Zhong, Chao Zhao, Anhua Chen, and Tianyu Zhang. A uav- based machine vision method for bridge crack recognition and width quantification through hybrid feature learning. Construction and Building Materials, 299:123896, 2021. [12] Wei Ding, Han Yang, Ke Yu, and Jiangpeng Shu. Crack detection and quantifica-tion for concrete structures using uav and transformer. Automation in Construction,152:104929, 2023.98 doi:10.6342/NTU202403413 [13] Xinyu He, Zhiwen Tang, Yubao Deng, Guoxiong Zhou, Yanfeng Wang, and Liujun Li. Uav-based road crack object-detection algorithm. Automation in Construction, 154:105014, 2023. [14] Ruoxian Li, Jiayong Yu, Feng Li, Ruitao Yang, Yudong Wang, and Zhihao Peng. Automatic bridge crack detection using unmanned aerial vehicle and faster r-cnn. Construction and Building Materials, 362:129659, 2023. [15] 范淳皓. 基於深度強化學習神經網路之自動化裂縫分割與偵測. Master’s thesis, Jan 2024. [16] Yahui Liu, Jian Yao, Xiaohu Lu, Renping Xie, and Li Li. Deepcrack: A deep hi-erarchical feature learning architecture for crack segmentation. Neurocomputing, 338:139–153, 2019. [17] Yao Yao, Shue-Ting Ellen Tung, and Branko Glisic. Crack detection and char-acterization techniques—an overview. Structural Control and Health Monitoring,21(12):1387–1413, 2014. [18] P Iakubovskii. Segmentation models pytorch, 2019, github repository, github. URL https://github. com/qubvel/segmentation_models. pytorch. [19] Edwin Salcedo, Mona Jaber, and Jesús Requena Carrión. A novel road maintenance prioritisation system based on computer vision and crowdsourced reporting. Journalof Sensor and Actuator Networks, 11(1), 2022. [20] Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional net-works for biomedical image segmentation. In Nassir Navab, Joachim Hornegger, William M. Wells, and Alejandro F. Frangi, editors, Medical Image Computing 99 doi:10.6342/NTU202403413 and Computer-Assisted Intervention – MICCAI 2015, pages 234–241, Cham, 2015. Springer International Publishing. [21] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016. [22] P Iakubovskii. Segmentation models pytorch, 2019, github repository, github. URL https://github. com/qubvel/segmentation_models. pytorch. [23] Christopher JCH Watkins and Peter Dayan. Q-learning. Machine learning, 8:279–292, 1992. [24] Richard Bellman. Dynamic programming. science, 153(3731):34–37, 1966. [25] Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra, and Martin Riedmiller. Playing atari with deep rein- forcement learning. arXiv preprint arXiv:1312.5602, 2013. [26] Hado van Hasselt, Arthur Guez, and David Silver. Deep reinforcement learning with double q-learning. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1), Mar. 2016. [27] Sebastián Valladares, Mayerly Toscano, Rodrigo Tufiño, Paulina Morillo, and Diego Vallejo-Huanga. Performance evaluation of the nvidia jetson nano through a real- time machine learning application. In Intelligent Human Systems Integration 2021: Proceedings of the 4th International Conference on Intelligent Human Systems Integration (IHSI 2021): Integrating People and Intelligent Systems, February 22-24, 2021, Palermo, Italy, pages 343–349. Springer, 2021.100 doi:10.6342/NTU202403413 [28] Eduardo Assunção, Pedro D Gaspar, Ricardo Mesquita, Maria P Simões, Khadijeh Alibabaei, André Veiros, and Hugo Proença. Real-time weed control application using a jetson nano edge device and a spray mechanism. Remote Sensing, 14(17):4217,2022. [29] Nathan Koenig and Andrew Howard. Design and use paradigms for gazebo, an open-source multi-robot simulator. In 2004 IEEE/RSJ international conference on intelligent robots and systems (IROS)(IEEE Cat. No. 04CH37566), volume 3, pages 2149–2154. Ieee, 2004. [30] Johannes Meyer, Alexander Sendobry, Stefan Kohlbrecher, Uwe Klingauf, and Oskar Von Stryk. Comprehensive simulation of quadrotor uavs using ros and gazebo. In Simulation, Modeling, and Programming for Autonomous Robots: Third International Conference, SIMPAR 2012, Tsukuba, Japan, November 5-8, 2012. Proceedings 3, pages 400–411. Springer, 2012. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/94063 | - |
| dc.description.abstract | 在鋼筋混凝土建築中，裂縫不僅是結構健康與安全的重要指標，更是地震後評估與補強工作中至關重要的信息來源。地震發生時，結構工程師需迅速評估裂縫的尺寸、形態及位置，以判斷是否存在即時的結構危害，並制定相應的補強策略。因此，裂縫檢測在現代社會中具有重要意義，這項技術不僅影響到建築物的長期穩定性，也直接關乎公眾的安全與福祉。近年來，隨著硬體技術的迅速進步及人力成本的提高，無人機技術開始被廣泛應用於裂縫檢測領域。儘管學術界已有大量相關研究，但許多應用仍限於無人機作為資料收集平台，未能實現裂縫的自動捕捉及自主飛行。本研究針對此問題，致力於開發一種新型無人機飛行系統，專為實現裂縫的自動捕捉和強化自主飛行能力而設計。首先，我們採用深度強化學習方法對模型進行精細訓練，使其能夠基於局部裂縫特徵有效判斷最佳飛行路徑。其次，為實現無人機的自主飛行，我們運用MAVLINK無人機通信協議編寫了專屬的飛行控制指令，以確保系統在各種環境條件下能夠靈活適應並執行任務。同時，我們在系統設計中融合了先進的軟體與硬體技術，保證無人機在飛行過程中能夠即時進行邊緣運算，實現高效的全自動操作。實驗結果表明，該系統成功捕捉並分析了62%的可見裂縫，檢測準確率為86%。這些發現突顯了該系統在提高結構檢測的可靠性和效率方面的潛力。 | zh_TW |
| dc.description.abstract | In reinforced concrete structures, cracks serve not only as crucial indicators of structural health and safety but also as vital sources of information for post-earthquake assessment and reinforcement planning. During seismic events, structural engineers must swiftly assess the size, morphology, and locations of cracks to determine immediate structural hazards and devise appropriate reinforcement strategies. Hence, crack detection holds significant importance in modern society, impacting both the long-term stability of buildings and public safety and well-being directly. In recent years, with rapid advancements in hardware technology and the increasing costs of human labor, unmanned aerial vehicle (UAV) technology has found widespread application in the field of crack detection. Despite extensive academic research, many applications still confine UAVs to data collection platforms, failing to achieve automated crack detection and autonomous flight. Addressing this gap, this study focuses on developing a novel UAV flight system designed specifically for automated crack detection and enhanced autonomous flight capabilities. Firstly, we employ a deep reinforcement learning approach to finely train the model, enabling it to effectively determine optimal flight paths based on local crack features. Secondly, to achieve autonomous flight, we utilize MAVLink UAV communication protocol to develop tailored flight control commands, ensuring the system's adaptability and mission execution across diverse environmental conditions. Concurrently, our system design integrates advanced software and hardware technologies to facilitate real-time edge computing during UAV flight, thereby achieving efficient fully automated operations. Experimental results indicate that the system successfully captured and analyzed 62% of visible cracks, with a detection accuracy rate of 86%. These findings underscore the system's potential in improving the reliability and efficiency of structural inspections. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2024-08-14T16:30:22Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2024-08-14T16:30:22Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | Acknowledgements i
摘要 iii Abstract v Contents vii List of Figures xi List of Tables xvii Chapter 1 Introduction 1 1.1 Research Background and Motivation 1 1.2 Literature Review 3 1.3 Research Objectives 6 1.4 Contribution and Scopes 7 Chapter 2 Methodology 9 2.1 Model Training 10 2.1.1 Datasets 11 2.1.2 Segmentation 13 2.1.2.1 U-Net 14 2.1.2.2 ResNet 15 2.1.2.3 U-Net-ResNet34 17 2.1.2.4 Segmentation Result 18 2.1.3 Deep Reinforcement Learning 19 2.1.3.1 Deep Reinforcement Learning 20 2.1.3.2 Q-learning 22 2.1.3.3 Deep Q-learning 24 2.1.3.4 Double Deep Q-learning 26 2.1.4 Training Result 28 2.1.5 UAV Platform Implementation 31 2.1.5.1 UAV Image Input 32 2.1.5.2 Comparison of Offboard and Onboard 34 2.2 Flight Control Commands 36 2.2.1 Mavlink Protocol 37 2.2.2 MAVLink Task Commands 38 2.2.2.1 Coordinate Frame 40 2.2.2.2 Bitmask 41 2.3 Integrated Platform 43 2.3.1 Software 43 2.3.1.1 Ubuntu version 43 2.3.1.2 Flowchart 43 2.3.2 Hardware 45 2.3.2.1 Jetson Nano 45 2.3.2.2 AXM-9109 48 2.3.3 Firmware integration 51 2.3.3.1 Image Transmission Issue 51 2.3.3.2 Unstable Flight Control Transmission 54 2.3.3.3 Edge Computing Optimization in Firmware 56 2.4 Calculation of Crack Width in Real World 57 2.4.1 Formula Derivation 58 Chapter 3 Implementation 65 3.1 Simulation 65 3.1.1 Gazebo 66 3.1.2 Testing Architecture 67 3.1.3 Simulation Test 69 3.2 Real Flight 73 3.2.1 Camera Orientation for Experiment 73 3.2.2 Flight Altitude and Camera Zoom 74 3.2.3 Real Flight Test 77 3.2.3.1 Test A 77 3.2.3.2 Test B 82 Chapter 4 Result and discussion 87 4.1 Flight Result 87 4.2 Validation of Crack Width in Real World 89 4.3 Limitation 90 Chapter 5 Conclusion 93 5.1 Summary 93 5.2 Future work 94 References 97 | - |
| dc.language.iso | en | - |
| dc.subject | 無人機 | zh_TW |
| dc.subject | 裂縫檢測 | zh_TW |
| dc.subject | 自主飛行 | zh_TW |
| dc.subject | 邊緣計算 | zh_TW |
| dc.subject | 深度強化學習 | zh_TW |
| dc.subject | MAVLINK | en |
| dc.subject | Autonomous Flight | en |
| dc.subject | Edge Computing | en |
| dc.subject | Unmanned Aerial Vehicle | en |
| dc.subject | Deep Reinforcement Learning | en |
| dc.subject | Crack Detection | en |
| dc.title | 以裂縫自動探索為導向之無人機整合飛控系統研發 | zh_TW |
| dc.title | Development of an Integrated UAV Flight Control System for Automated Crack Detection and Exploration | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 112-2 | - |
| dc.description.degree | 碩士 | - |
| dc.contributor.oralexamcommittee | 韓仁毓;林子剛; 葉芳耀 | zh_TW |
| dc.contributor.oralexamcommittee | Jen-Yu Han;Tzu-Kang Lin;Fang-Yao Yeh | en |
| dc.subject.keyword | 裂縫檢測,無人機,深度強化學習,邊緣計算,自主飛行, | zh_TW |
| dc.subject.keyword | Crack Detection,Unmanned Aerial Vehicle,Deep Reinforcement Learning,MAVLINK,Edge Computing,Autonomous Flight, | en |
| dc.relation.page | 101 | - |
| dc.identifier.doi | 10.6342/NTU202403413 | - |
| dc.rights.note | 同意授權(全球公開) | - |
| dc.date.accepted | 2024-08-13 | - |
| dc.contributor.author-college | 工學院 | - |
| dc.contributor.author-dept | 土木工程學系 | - |
| 顯示於系所單位： | 土木工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-112-2.pdf | 29.49 MB | Adobe PDF | 檢視/開啟 |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。