簡易檢索 / 詳目顯示

回結果列表
研究生： 宮維澤
Koung, Wei-Tse
論文名稱： 應用類神經網路改善室內無人機超寬頻定位技術之準確性
Enhancing Ultra-Wideband Positioning for Indoor UAVs Using Artificial Neural Network
指導教授： 陳偉良
Chan, Woei-Leong
學位類別： 碩士
Master
系所名稱： 工學院 - 民航研究所
Institute of Civil Aviation
論文出版年： 2024
畢業學年度： 113
語文別： 英文
論文頁數： 54
中文關鍵詞： 室內定位 、超寬頻 、類神經網路
外文關鍵詞： Indoor Positioning, UWB, Neural Network
相關次數： 點閱：80  下載：19 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 衛星定位系統 (GNSS) 在車輛導航、智慧手機定位服務、航空航海導航、農業、搜救、運動追蹤、地質探勘及科學研究等多個領域都有廣泛運用。它在提高方便性和安全性方面發揮了關鍵的作用。然而，由於衛星定位無法對於室內定位提供穩定的定位服務，因此在室內環境需要採用其他定位方式來取代GNSS。

    隨著農業科技的進步，機器人已廣泛應用於農業生產，但在室內環境中，缺乏GNSS定位仍然是一個挑戰。為了克服這個問題，本研究使用了UWB(Ultra Wide Band)超寬頻技術進行室內定位和導航。面對UWB室內定位遭遇的天線輻射場型造成定位不精確的情況，利用類神經網路估測UWB 誤差，以增強原有演算法的可靠性。這一研究成果使得無人載具能更順利地按照預設航點進行飛行，降低了失控和墜毀的風險，同時提高了作業安全性。這種創新應用為農業生產帶來了更大的便利性和效益。

    The Global Navigation Satellite System (GNSS) is widely utilized in various fields such as vehicle navigation, smartphone location services, aviation and maritime navigation, agriculture, search and rescue operations, sports tracking, geological exploration, and scientific research. It plays a crucial role in enhancing convenience and safety. However, due to the inability of satellite positioning to provide accurate services for indoor environments, alternative positioning methods are required to replace GNSS in indoor environments.

    With the advancement of agricultural technology, robots have been extensively used in agricultural production. However, the lack of GNSS navigation remains a challenge in indoor environments. To overcome this issue, this thesis focuses on the Ultra-Wideband (UWB) technology for indoor positioning and navigation. Faced with the possibility of antenna radiation pattern induced biases, artificial neural network (ANN) was utilized to estimate the bias to enhances the reliability of existing algorithms. This research outcome enables unmanned aerial vehicles to fly more smoothly according to predefined waypoints, reducing the risk of loss of control and crashes while increasing operational safety. This innovative application brings greater convenience and efficiency to agricultural production.

    中文摘要 i Abstract ii Acknowledgements iii Table of Contents iv List of Figures v List of Table viii I. Introduction 1 1.1 Research Background 1 1.2 Literature Review 2 1.3 Motivations and Objectives 10 1.4 Thesis Organization 10 II. Methodology 11 2.1 System Overview and Setup 11 2.2 UWB TWR Measurement Ranging Bias 15 2.3 Interim Summary 18 III. Data Processing 19 3.1 Chi-Square Goodness of Fit 19 3.2 Root Mean Square Deviation 20 3.3 Use Two Methods to Process Raw Data 21 3.4 Interim Summary 26 IV. Model Training and Discussion 27 4.1 Bayesian Regularization of Artificial Neural Networks 27 4.2 Training of Feed-Forward Neural Network 28 4.3 Training Results 31 4.4 Verify the Artificial Neural Network Model 33 4.5 Deploy to the Crazyflie Firmware 35 4.6 Discussion 37 V. Conclusion 39 References 40 Appendix 42 Appendix A. Outlier Bias Example (0.8m) 42 Appendix B. Outlier Bias Example (1.4m) 43 Appendix C. Outlier Bias Example (1.8m) 44

    [1] Chi G, Yang Z, Xu J, Wu C, Zhang J, Liang J, Liu Y, “Wi-drone: Wi-Fi-Based 6-DoF Tracking for Indoor Drone Flight Control,” Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services (MobiSys '22), 2022, pp.56-68
    [2] Niculescu V, Palossi D, Magno M, Benini L, “Energy-Efficient, Precise UWB-Based 3-D Localization of Sensor Nodes with a Nano-UAV,” IEEE Internet of Things Journal, Vol. 10, No. 7, 2023, pp. 5760-5777
    [3] Li C, Tanghe E, Suanet P, Plets D, Hoebeke J, De Poorter E; Joseph W, “ReLoc 2.0: UHF-RFID Relative Localization for Drone-Based Inventory Management,” IEEE Transactions on Instrumentation and Measurement, Vol. 70, 2021, pp. 1-13
    [4] L. Chang, K. Li, and B. Hu, "Huber’s M-estimation-based process uncertainty robust filter for integrated INS/GPS," IEEE Sensors Journal, vol. 15, no. 6, pp. 3367-3374, 2015.
    [5] W. Zhao, J. Panerati, and A. P. Schoellig, "Learning-based bias correction for time difference of arrival ultra-wideband localization of resource-constrained mobile robots," IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 3639-3646, 2021.
    [6] W.-L. C. W.-T. Koung, R.-H. Lee, and C.-C. Chen, "Ranging Bias Estimation of Ultra-wideband Localization System using Bayesian Regularization Backpropagation Neural Network for Indoor Agricultural Drone Positioning."Journal of Aeronautics, Astronautics and Aviation, submitted for publication.
    [7] J. Schmidhuber, "Deep learning in neural networks: An overview," Neural networks, vol. 61, pp. 85-117, 2015.
    [8] "DWM1000 3.5 - 6.5 GHz Ultra-Wideband (UWB) Transceiver Module Documents, NLOS Operation and Optimizations to Improve Performance."
    [9] "Bitcraze (https://www.bitcraze.io/)."
    [10] "VICON (https://www.vicon.com/)."
    [11] Burden F, Winkler D. Bayesian regularization of neural networks. Methods Mol Biol. 2008;458:25-44. doi: 10.1007/978-1-60327-101-1_3. PMID: 19065804.
    [12] Vogl, T. P., et al. ‘Accelerating the Convergence of the Back-Propagation Method’. Biological Cybernetics, vol. 59, no. 4–5, Sept. 1988, pp. 257–63. DOI.org (Crossref), doi:10.1007/BF00332914.
    [13] Lohan, E. S., & Auer, S. (2013). Ultra-wideband based indoor positioning systems: a survey. In Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) (pp. 2009-2017). The Institute of Navigation.
    [14] Patwari, N., Kasera, S. K., & Kyperountas, S. (2005). High-precision time-of-arrival-based localization in ad hoc networks with nonsynchronized nodes. IEEE Transactions on Mobile Computing, 4(5), 458-473.
    [15] Pozzi, M., Guerrieri, A., Zanca, G., & Thoma, R. S. (2010). Energy-efficient cooperative positioning in UWB sensor networks. IEEE Transactions on Vehicular Technology, 59(9), 4338-4352.
    [16] Le, L. A., & Hwang, S. H. (2016). Performance evaluation of indoor positioning system based on UWB ranging technique. IEEE Access, 4, 8257-8268.

    下載圖示 校內：立即公開
    校外：立即公開
    QR CODE