簡易檢索 / 詳目顯示

回結果列表
研究生： 黃詩婷
Huang, Shih-Ting
論文名稱： 基於時間卷積網路進行行人航位推算法之步伐長度估計研究
The Study of Step Length Estimation of Pedestrian Dead Reckoning Based on Temporal Convolution Network
指導教授： 黃之浩
Huang, Scott Chih-Hao
口試委員: 高榮駿
Kao, Jung-Chun
李端興
Lee, Duan-Shin
葉弼群
Yeh, Bih-Chyun
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 通訊工程研究所
Communications Engineering
論文出版年： 2021
畢業學年度： 109
語文別： 中文
論文頁數： 66
中文關鍵詞： 室內定位 、多模式行人航位推算系統 、機器學習 、神經網路模型 、非線性特徵轉換 、自適應式步長演算法
外文關鍵詞： Indoor Localization, Multimode Pedestrian Dead Reckoning System, Machine Learning, Neural Network Model, Nonlinear Feature Transformation, Adaptive Step Length Algorithm
相關次數： 點閱：2  下載：0 
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

    • 近年來，人們對於室內定位的需求層面日益擴大，現今室外定位主流的GPS系統因其應用於室內易受遮蔽訊號以及多路徑效應影響，使得定位成效不彰，鑒於此現象，目前眾多研究致力於行人航位推算(Pedestrian Dead Reckoning, PDR)系統的分析，以降低定位誤差為核心目標進行探討。
    目前常見的行人航位推算研究大多數固定單一手機姿態去分析，限制行人以手持模式下紀錄數據，若隨意改變其設備姿態將造成誤差增大，本文基於上述缺失去改進，應用三種常見的設備攜帶模式進行分析，實現一套多模式行人航位推算系統。架構中主要使用三軸線性加速度計(Three-axis Accelerometer)、三軸陀螺儀(Three-axis Gyroscope)、重力感測器(Gravity Sensor)以及旋轉向量感測器(Rotation Vector Sensor)四個慣性感測單元(Inertial measurement unit, IMU)作為研究的數據來源，基於不同模式在步伐檢測區塊提出各自的分析方法，並使用機器學習演算法進行模式的分類輔助其運作，而在步伐長度估算區塊中取代傳統的經驗模型，構建神經網路模型運行迴歸分析，並實現一種非線性特徵轉換方式以增進模型的成效，能夠適應不同使用者的行走習慣，屬於一種自適應式步長演算法。實驗結果顯示，本文提出的架構在各區塊皆有顯著的進展，並且能夠有效降低整體的定位誤差。

    The demand for indoor localization has expanded greatly in recent years. When the pedestrians are in the building, the received signal will be blocked and affect the performance of GPS positioning. As a result, the numerous researches are devoted to the analysis of pedestrian dead reckoning system.
    In this article, a multimode pedestrian dead reckoning system is proposed based on different device placements to improve positioning error. We have used the data which includes accelerometer, gyroscope, gravity sensor and rotation vector sensor in the framework.
    In the step detection block, we propose different analysis methods based on different phone poses and work successfully with machine learning classifier. Furthermore, in the step length estimation block, we haven’t used an empirical model. Instead, we have built a neural network sequence model for regression analysis. In the prior works, the features are extracted from the raw acceleration and angular velocity signal. However, the non-linear feature transformation method is adopted to generate tree features in this article. We design an adaptive step length algorithm considering gait diversity. Experimental results show that the proposed architecture has made much progress in each block, and can effectively reduce the overall positioning error.

    第一章、緒論 1 1.1 研究背景 1 1.2 研究動機與目的 3 1.3 論文架構 4 第二章、相關研究 5 2.1 行人航位推算(Pedestrian Dead Reckoning, PDR) 5 2.1.1 概述 5 2.1.2 步伐檢測 (Step detection) 6 2.1.3 步伐長度估算 (Step Length Estimation) 9 2.1.4 方位估計 (Heading Estimation) 11 2.2 神經網路模型 11 2.2.1 雙向長短期記憶神經網路 (Bi-directional LSTM) 11 2.2.2 時間卷積神經網路(Temporal Convolution Network, TCN) 13 2.3 特徵轉換 (Feature Transformation) 16 2.4 巴特沃斯濾波 18 第三章、空間座標系統 20 3.1 座標系定義 20 3.1.1 地心地固座標系(Earth-Centered Earth-Fixed Frame, ECEF) 20 3.1.2 導航坐標系(Navigation Frame) 21 3.1.3 附體坐標系(Body Frame) 21 3.2 座標系旋轉 22 3.2.1 方向餘弦法 22 3.2.1 四元數法 24 3.2.3 尤拉角法 26 3.3 座標系轉換 27 第四章、系統架構 29 4.1 概述 29 4.2 設備模式 29 4.3 改進的行人航位推算 31 4.3.1 步伐檢測 31 4.3.2 步伐長度估算 36 4.3.3 方位估計 41 第五章、模擬與結果 43 5.1 資料集 43 5.2 模擬結果 43 5.2.1 設備模式 43 5.2.2 步伐檢測 48 5.2.3 步伐長度估算 51 5.2.4 方位估計 57 第六章、結論與未來工作 63 參考文獻 64

    [1] Pei Ling, Ruizhi Chen, Jingbin Liu, Tomi Tenhunen, Heidi Kuusniemi, and Yuwei Chen, "Inquiry-based bluetooth indoor positioning via rssi probability distributions", IEEE Second International Conference on Advances in Satellite and Space Communications, 151-156, 2010.
    [2] Beom Ju Shin, Kwang Won Lee, Sun Ho Choi, Joo Yeon Kim, Woo Jin Lee, and Hyung Seok Kim, "Indoor WiFi positioning system for Android-based smartphone", IEEE International conference on information and communication technology convergence, 319-320, 2010.
    [3] Antonio Ramón Jiménez Ruiz, Fernando Seco Granja, José Carlos Prieto Honorato, and Jorge I. Guevara Rosas, "Accurate pedestrian indoor navigation by tightly coupling foot-mounted IMU and RFID measurements", IEEE Transactions on Instrumentation and Measurement, 61: 178-189, 2011.
    [4] Changhao Chen, Xiaoxuan Lu, Andrew Markham, and Niki Trigoni, "Ionet: Learning to cure the curse of drift in inertial odometry", AAAI Conference on Artificial Intelligence, 6468-6476, 2018.
    [5] Itzik Klein, Yuval Solaz, and Guy Ohayon, "Pedestrian dead reckoning with smartphone mode recognition", IEEE Sensors Journal, 18: 7577-7584, 2018.
    [6] Qinglin Tian, Zoran Salcic, Kevin Wang, and Yun Pan, "A multi-mode dead reckoning system for pedestrian tracking using smartphones", IEEE Sensors Journal, 16: 2079-2093, 2015.
    [7] Zhenghua Chen, Han Zou, Hao Jiang, Qingchang Zhu, Yeng Chai Soh, and Lihua Xie, "Fusion of WiFi smartphone sensors and landmarks using the Kalman filter for indoor localization", Sensors, 15: 715-732, 2015.
    [8] Inge Bylemans, Maarten Weyn, and Martin Klepal, "Mobile phone-based displacement estimation for opportunistic localisation systems", IEEE Third International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies, 113-118, 2009.
    [9] Yunye Jin, Hong Song Toh, Wee Seng Soh, and Wai Choong Wong, "A Robust Dead-Reckoning Pedestrian Tracking System with Low Cost Sensors", IEEE International Conference on Pervasive Computing and Communications, 222-230, 2011.
    [10] Lei Fang, Panos Antsaklis, Luis Antonio Montestruque, Brett McMickell, Michael Lemmon, Yashan Sun, Hui Fang, Ioannis Koutroulis, Martin Haenggi, and Min Xie, "Design of a wireless assisted pedestrian dead reckoning system-the NavMote experience", IEEE Transactions on Instrumentation and Measurement, 54: 2342-2358, 2005.
    [11] Beomju Shin, Chulki Kim, Jaehun Kim, Seok Lee, Changdon Kee, Hyoung Seok Kim, and Taikjin Lee, "Motion recognition-based 3D pedestrian navigation system using smartphone", IEEE Sensors Journal, 16: 6977-6989. 2016.
    [12] Pragun Goyal, Vinay J. Ribeiro, Huzur Saran, and Anshul Kumar, "Strap-down pedestrian dead-reckoning system", IEEE International Conference on Indoor Positioning and Indoor Navigation, 1-7, 2011.
    [13] Seung Hyuck Shin, Min Su Lee, Chan Gook Park, and Hyun Su Hong, "Pedestrian dead reckoning system with phone location awareness algorithm", IEEE/ION Position, Location and Navigation Symposium, 97-101, 2010.
    [14] Seung Hyuck Shin, and Chan Gook Park, "Adaptive step length estimation algorithm using optimal parameters and movement status awareness", Medical engineering and physics, 33: 1064-71, 2011.
    [15] Valérie Renaudin, Melania Susi, and Gérard Lachapelle, "Step length estimation using handheld inertial sensors", Sensors, 12: 8507-8025, 2012.
    [16] Youngwoo Kim, Odongo Steven Eyobu, and Dong Seog Han, "ANN-based stride detection using smartphones for Pedestrian dead reckoning", IEEE International Conference on Consumer Electronics, 1-2, 2018.
    [17] Harvey Weinberg, "Using the ADXL202 in pedometer and personal navigation applications", Analog Devices AN-602 application note, 2: 1-6, 2002.
    [18] Jeong Won Kim, Han Jin Jang, Dong Hwan Hwang, and Chansik Park, "A step, stride and heading determination for the pedestrian navigation system", Journal of Global Positioning Systems, 3: 273-279, 2004.
    [19] Jim Scarlett, "Enhancing the performance of pedometers using a single accelerometer", Analog Devices application note, 2007.
    [20] Itzik Klein, Yuval Solaz, and Guy Ohayon, "Pedestrian dead reckoning with smartphone mode recognition", IEEE Sensors Journal, 18: 7577-7584, 2018.
    [21] Qinglin Tian, Zoran Salcic, Kevin Wang, and Yun Pan, "An enhanced pedestrian dead reckoning approach for pedestrian tracking using smartphones", IEEE Tenth International Conference on Intelligent Sensors, Sensor Networks and Information Processing, 1-6, 2015.
    [22] Boyuan Wang, Xuelin Liu, Baoguo Yu, Ruicai Jia, and Xingli Gan, "Pedestrian dead reckoning based on motion mode recognition using a smartphone", Sensors, 18: 1811, 2018.
    [23] Jin Shyan Lee, and Shih Min Huang, "An experimental heuristic approach to multi-pose pedestrian dead reckoning without using magnetometers for indoor localization", IEEE Sensors Journal, 19: 9532-9542, 2019.
    [24] Xiaokun Yang, Baoqi Huang, and Qing Miao, "A step-wise algorithm for heading estimation via a smartphone", Chinese Control and Decision Conference, 4598-4602, 2016.
    [25] Wonho Kang, Seongho Nam, Youngnam Han, and Sookjin Lee, "Improved heading estimation for smartphone-based indoor positioning systems", IEEE 23rd International Symposium on Personal, Indoor and Mobile Radio Communications, 2449-2453, 2012.
    [26] Shaojie Bai, J Zico Kolter, and Vladlen Koltun, "An empirical evaluation of generic convolutional and recurrent networks for sequence modeling", 2018.
    [27] Xinran He, Junfeng Pan, Ou Jin, Tianbing Xu, Bo Liu, Tao Xu, Yanxin Shi, Antoine Atallah, Ralf Herbrich, and Stuart Bowers, "Practical lessons from predicting clicks on ads at []book", Proceedings of the Eighth International Workshop on Data Mining for Online Advertising, 1-9, 2014.
    [28] Hang Yan, Qi Shan, and Yasutaka Furukawa, "RIDI: Robust IMU double integration", Proceedings of the European Conference on Computer Vision, 621-636, 2018.
    [29] Dominik Gusenbauer, Carsten Isert, and Jens Krösche, "Self-contained indoor positioning on off-the-shelf mobile devices", International Conference on Indoor Positioning and Indoor Navigation, 1-9, 2010.
    [30] Jiuchao Qian, Jiabin Ma, Rendong Ying, Peilin Liu, and Ling Pei, "An improved indoor localization method using smartphone inertial sensors", International Conference on Indoor Positioning and Indoor Navigation, 1-7, 2013.

    QR CODE