藉助慣性感測單元來輔助之運動健身系統的研究與開發

簡易檢索 / 詳目顯示

回結果列表

研究生：	林奕廷 Lin, Yi-Ting
論文名稱：	藉助慣性感測單元來輔助之運動健身系統的研究與開發 An IMU-aided Fitness System
指導教授：	王俊堯 Wang, Chun-Yao
口試委員:	邱文信 Chiu, Wen-Hsin 李昀儒 Lee, Yun-Ju
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 資訊工程學系 Computer Science
論文出版年：	2020
畢業學年度：	108
語文別：	英文
論文頁數：	29
中文關鍵詞：	慣性感測單元、運動健身
外文關鍵詞：	IMU, Fitness
相關次數：	點閱：2 下載：0
分享至:	分享至facebook 分享至twitter

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

在此研究中，主要針對四種健身動作進行動作辨識以及動作錯誤判別，包含負重深蹲、橋式運動、棒式運動、及仰臥抬腿。並且，藉由慣性測量單元(Inertial measurement unit)之輔助來開發一款Android APP，使得系統可回報運動動作之正確性、呈現過往運動表現、及上傳運動數據至個人雲端。最後，實驗證明所有辨識項目準確率可達97%以上，以及平均的絕對平均誤差在2°以內。

In this work, we present an IMU-aided fitness system for users conducting exercises. The system utilizes three wearable 9-axis IMU sensors to detect four kinds of exercises including squat, bridge exercise, plank exercise, and double leg raise. The sensors are also connected to a smartphone via Bluetooth, and an Android APP is designed for users. The system can instantaneously report the correctness of exercises, show the statistics of workouts, and record the data in the cloud for references. The experimental results demonstrate that the average recognition rates are greater than 97%, and the average absolute mean error in movement angle and swing angle are less than 2°.

1 Introduction 1
2 IMU Sensors 5
3 Method 7
3.1 Squat with barbells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2 Bridge exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.3 Plank exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.4 Double leg raise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4 Android Application 18
4.1 The APP page for squat . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.2 The APP page for bridge exercise . . . . . . . . . . . . . . . . . . . . 19
4.3 The APP page for plank exercise . . . . . . . . . . . . . . . . . . . . 20
4.4 The APP page for double leg raise . . . . . . . . . . . . . . . . . . . 20
4.5 The statistics of workouts . . . . . . . . . . . . . . . . . . . . . . . . 21
4.6 Real-time audio feedback . . . . . . . . . . . . . . . . . . . . . . . . . 22
5 Experimental Results 23
5.1 Recognition rate in squat . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.2 Recognition rate in bridge exercise . . . . . . . . . . . . . . . . . . . 24
5.3 Recognition rate in plank exercise . . . . . . . . . . . . . . . . . . . . 24
5.4 Recognition rate in double leg raise . . . . . . . . . . . . . . . . . . . 25
5.5 Angle calculation in bridge exercise . . . . . . . . . . . . . . . . . . . 25
5.6 Angle calculation in double leg raise . . . . . . . . . . . . . . . . . . . 25
6 Conclusions 27
Bibliography 28

List of Tables
3.1 The conditions for detecting squats. “S” represents the Setting value.
“A.V.” represents Angular Velocity. . . . . . . . . . . . . . . . . . . . 11
3.2 The conditions for detecting bridge exercises. . . . . . . . . . . . . . 13
3.3 The conditions for detecting plank exercises. . . . . . . . . . . . . . . 15
3.4 The conditions for detecting double leg raises. . . . . . . . . . . . . . 17
5.1 Experimental results. . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

List of Figures
2.1 The setting of the 9-axis IMU sensor. (a) The demonstration of Euler
angle of the sensor. (b) The demonstration of positive and negative
intervals of the Roll angle of the sensor. . . . . . . . . . . . . . . . . . 6
3.1 The wearable devices on wrist and knee. . . . . . . . . . . . . . . . . 8
3.2 Three calibration motions for squat. (a) The posture of stand. (b)
The posture of pre-squat (c) The posture of squat without holding
barbells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.3 Good morning squat demonstration (down). (a) Posture of pre-squat.
(b) Posture of normal squat. (c) Posture of good morning squat and
the injured position. . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.4 Good morning squat demonstration (up). (a) Posture of normal
squat. (b) Posture of good morning squat and the injured position.
(c) Posture of pre-squat. . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.5 Hip joint deactivation demonstration. (a) Posture of hip joint activation. (b) Posture of hip joint deactivation. . . . . . . . . . . . . . . . 10
3.6 The wearable devices on the chest and waist. . . . . . . . . . . . . . . 11
3.7 Two calibration motions for bridge exercise. (a) Posture of flat back.
(b) Posture of bridge. . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.8 The demonstration of the posture of neutral spine and non-neutral
spine for bridge exercise. (a) The posture of neutral spine. (b) The
posture of non-neutral spine. . . . . . . . . . . . . . . . . . . . . . . . 12
3.9 The wearable device on waist. . . . . . . . . . . . . . . . . . . . . . . 13
3.10 The calibration motion and the adjustment to sensor for plank exercise. (a) The calibration motion. (b) The adjustment to the sensor. . 14
3.11 The demonstration of the postures for standard and deviation of plank
exercise. (a) The posture of standard plank exercise. (b) The posture
of up deviation. (c) The posture of down deviation. . . . . . . . . . . 15
3.12 The wearable devices on knee and lower back. . . . . . . . . . . . . . 16
3.13 Three calibration motions for double leg raise. (a) Posture of raise.
(b) Posture of drop. (c) Posture of flat back. . . . . . . . . . . . . . . 17
4.1 The APP pages for squat and bridge exercise. (a) The page for squat.
(b) The page for bridge exercise. . . . . . . . . . . . . . . . . . . . . . 19
4.2 The APP pages for plank exercise and double leg raise. (a) The page
for plank exercise. (b) The page for double leg raise. . . . . . . . . . 21
4.3 The demonstration of statistics page. . . . . . . . . . . . . . . . . . . 21
                                

Bibliography
[1] M. S. Ashapkina, A. V. Alpatov, V. A. Sablina and A. A. Chekushin, “Quantitative Parameters Developing for Estimation of Rehabilitation Exercising on The Basis of IMU Sensor Device,” Mediterranean Conference on Embedded Computing, 2018, pp. 541-544.
[2] C. Crema, A. Depari, A. Flammini, E. Sisinni, T. Haslwanter and S. Salzmann, “IMU-based Solution for Automatic Detection and Classification of Exercises in The Fitness Scenario,” IEEE Sensors Applications Symposium, 2017, pp. 1-6.
[3] C. J. Chen, Y. T. Lin, C. C. Lin, Y. C. Chen, Y. J. Lee, and C. Y. Wang, “Rehabilitation System for Limbs using IMUs,” IEEE International Symposium on Quality Electronic Design, 2020.
[4] Y. P. Chang, T. C. Wang, Y. J. Lee, C. C. Lin, Y. C. Chen, and C. Y. Wang, “A Smart Single-Sensor Device for Instantaneously Monitoring Lower Limb Exercises,” IEEE International System-on-Chip Conference, 2019.
[5] P. Hausberger, A. Fernbach and W. Kastner, “IMU-based Smart Fitness Devices for Weight Training,” Annual Conference of the IEEE Industrial Electronics Society, 2016, pp. 5182-5189.
[6] D. Morris, T. S. Saponas, A. Guillory, I. Kelner, “RecoFit: Using a Wearable Sensor to Find, Recognize, and Count Repetitive Exercises,” Proceedings of Conference on Human Factors in Computing Systems, 2014, pp. 3225-3234.
[7] F. Wahjudi and F. J. Lin, “IMU-based Walking Workouts Recognition,” IEEE World Forum on Internet of Things, 2019, pp. 251-256.
[8] T. C. Wang, Y. P. Chang, C. J. Chen, Y. J. Lee, C. C. Lin, Y. C. Chen, and C. Y. Wang, “A Smart Knee Pad for Stride Count and Walking Distance Measurement via Knee Angle Calculation,” IEEE International Symposium on Quality Electronic Design, 2020.
[9] Fitbod, https://www.fitbod.me/.
[10] Gyro Gesture Control, http://www.gyro.com.tw/.
[11] Nike Training Club, https://www.nike.com/ntc-app/.
[12] Personal Trainer - Kaia, https://www.kaiahealth.com/.

簡易檢索 / 詳目顯示

相關論文