隨著羽球好手戴資穎和周天成等人在國際賽事嶄露頭角，羽球運動在台灣逐漸成為熱門的國民運動。為了解決傳統羽球訓練需要有人協助的缺點，本論文以物聯網的概念透過手機APP，整合了羽球發球機和球拍感測器，提出一個能和使用者互動的羽球訓練系統。
本系統由三個部分組成，分別是物聯網發球機、球拍感測器和智慧型手機。首先是物聯網發球機，將多台發球機經由Wi-Fi由手機控制發球，改善了傳統發球機發球時間間隔較長的問題。接著是搭載三軸加速度計和陀螺儀的球拍感測器，可以擷取使用者的揮拍動作，並透過藍牙傳至手機做球種辨識。在手機上，除了實現以序列最小優化演算法（SMO）為主的球種辨識演算法來即時地辨識出選手的擊球點和四個羽球球種之外，還能記錄選手揮拍的連續動作數據，作為日後離線分析的使用。本系統使用手機APP作為主要的控制工具與管理平台，開發的人機介面包括單台發球、隨機落點發球、排程發球、殺球訓練和防守訓練等預設的練習模式外，在互動功能中也能根據使用者球種辨識演算法的結果讓發球機做出適當的回擊。
在3位受試者中，離線分析的4種球種（長球、挑球、小球和殺球）辨識的個人模型和通用模型的平均準確度分別為98.33%和97.50%。在即時驗證的實驗中，辨識4種球種（長球、挑球、小球和殺球）的個人模型和通用模型的平均準確度分別為94.17%和97.08%。本論文建置的互動式羽球訓練系統，讓使用者得以借助數據化的資訊獲得訓練的結果分析，希望透過本系統的輔助可以提升訓練效率。
關鍵字：互動式、羽球訓練、手機APP、發球機、球種辨識

With the badminton players Tai Tzu Ying and Chou Tien Chen achieving good results in international competitions, badminton gradually becomes a popular sport in Taiwan. This thesis used the IoT concept to integrate the multiple IoT-based badminton service machines (BSMs) and a wireless racket sensor through the Mobile APP. It proposed a badminton training system that can interact with the user.
The system architecture includes three parts, IoT-based BSMs, a wireless racket sensor, and a Mobile APP. The multiple IoT-based BSMs can be controlled by our Mobile APP via Wi-Fi to serve shuttlecocks. The racket sensor can capture the user's swing motion by two inertial sensors and transmit it to the Mobile APP via Bluetooth Low Energy (BLE). The Mobile APP includes a stroke type recognition algorithm based on Sequential Minimal Optimization (SMO), which can detect the player's hitting point and recognize four kind of stroke types. This APP can also control multiple IoT-based BSMs with the preset practice modes. Besides, the IoT-based BSMs can also response and serve a shuttlecock appropriately based on the user's stroke type recognition algorithm.
Among the three subjects, the average accuracy of the personal model and the general model of 4 kind of stroke types (Clear, Lift, Net Shot, Smash) by offline analysis was 98.33% and 97.50%, respectively. In the real-time verification experiment, the average accuracy of the personal model and the general model of 4 kind of stroke types (Clear, Lift, Net Shot, Smash) was 94.17% and 97.08%, respectively. The proposed system allows users to obtain training results analysis with the help of digitized information, hoping to improve training efficiency through the assistance of this system.
Keywords： Interactive, Badminton Training, Mobile APP, Service Machine, Stroke Type Recognition

[1] T. Steels et al., "Badminton Activity Recognition Using Accelerometer Data," Sensors, 2020, vol. 20, no. 17, p. 4685. [Online]. Available: https://www.mdpi.com/1424-8220/20/17/4685.
[2] Badminton World Federation players, "Tai Tzu Ying," Accessed: June 8, 2021. [Online]. Available: https://bwfbadminton.com/player/61427/tai-tzu-ying.
[3] Badminton World Federation players, "Chou Tien Chen" Accessed: June 8, 2021. [Online]. Available: https://bwfbadminton.com/player/34810/chou-tien-chen.
[4] A. P. G. De Alwis, C. Dehikumbura, M. Konthawardana, T. D. Lalitharatne, and V. P. C. Dassanayake, "Design and Development of a Badminton Shuttlecock Feeding Machine to Reproduce Actual Badminton Shots," 2020 5th International Conference on Control and Robotics Engineering (ICCRE), 2020, pp. 73-77.
[5] M. U. Aslam, A. Bashir, W. U. Draz, H. M. Mahmood Raja, and Kamran, "Optimized Shuttlecock Propulsion Machine to Facilitate Badminton Training," 2019 International Conference on Electrical, Communication, and Computer Engineering (ICECCE), 2019, pp. 1-6.
[6] N. H. Kamarudin, W. S. Chung, B. Madon, and A. Khmag, "Internet of Things (IoT) with Mobile Educational Apps: A Review," 2019 IEEE 6th International Conference on Engineering Technologies and Applied Sciences (ICETAS), 2019, pp. 1-5.
[7] S. Mori, K. Tanaka, S. Nishikawa, R. Niiyama and Y. Kuniyoshi, "High-Speed and Lightweight Humanoid Robot Arm for a Skillful Badminton Robot," in IEEE Robotics and Automation Letters, vol. 3, no. 3, pp. 1727-1734, July 2018.
[8] S. Ramasinghe, K. G. Manosha Chathuramali, and R. Rodrigo, "Recognition of Badminton Strokes Using Dense Trajectories," 7th International Conference on Information and Automation for Sustainability, 2014, pp. 1-6.
[9] H. Y. Ting, K. S. Sim, and F. S. Abas, "Kinect-based Badminton Movement Recognition and Analysis System," International Journal of Computer Science in Sport, Jan. 2015, vol 14, pp. 25-41.
[10] N. P. Ju, D. R. Yu, T. U. İk, and W. C. Peng, "Trajectory-based Badminton Shots Detection," 2020 International Conference on Pervasive Artificial Intelligence (ICPAI), 2020, pp. 64-71.
[11] A. Jacob, W. N. Wan Zakaria, and M. R. B. Md Tomari, "Implementation of IMU Sensor for Elbow Movement Measurement of Badminton players," 2016 2nd IEEE International Symposium on Robotics and Manufacturing Automation (ROMA), 2016, pp. 1-6.
[12] Z. L. Wang, M. Guo, and C. Zhao, "Badminton Stroke Recognition Based on Body Sensor Networks," in IEEE Transactions on Human-Machine Systems, vol. 46, no. 5, pp. 769-775, Oct. 2016.
[13] M. A. I. Anik, M. Hassan, H. Mahmud, and M. K. Hasan, "Activity Recognition of a Badminton Game Through Accelerometer and Gyroscope," 2016 19th International Conference on Computer and Information Technology (ICCIT), 2016, pp. 213-217.
[14] Y. F. Wang, M. Chen, X. Y. Wang, R. H. M. Chan, and W. J. Li, "IoT for Next-Generation Racket Sports Training," in IEEE Internet of Things Journal, vol. 5, no. 6, pp. 4558-4566, Dec. 2018.
[15] Y. Wang, J. W. Ma, X. Li, and A. Zhong, "Hierarchical Multi-Classification for Sensor-based Badminton Activity Recognition," 2020 15th IEEE International Conference on Signal Processing (ICSP), 2020, pp. 371-375.
[16] Mega Sport international, "AS1063 Shuttlecock Service Machine", [Online]. Available: http://www.mega-sport.com.tw/en/product-description_1.html.
[17] C. A. Yuan, C. J. Tang, C. Li., J. J. Hu, and J. Peng, "A New Linear Approximate Clustering Algorithm Based Upon Sampling with Probability Distribution," 2005 International Conference on Machine Learning and Cybernetics, 2005, vol. 3, pp. 1518-1523.
[18] L. Y. Xu and F. Zhang, "Fractional Fourier Transform Estimation of Simple Randomly Sampled Signals," 2016 IEEE International Conference on Signal Processing, Communications and Computing (ICSPCC), 2016, pp. 1-5.
[19] C. Fan, P. Zhang, Y. Sun, C. Liu, and X. Shi, "Natural Disaster Information Statistics Study Based on Stratified Random Sampling Survey Statistical Methods," 2012 IEEE International Conference on Granular Computing, 2012, pp. 1-4.
[20] M. Peker, A. Arslan, B. Şen, F. V. Çelebi, and A. But, "A Novel Hybrid Method for Determining the Depth of Anesthesia Level: Combining ReliefF Feature Selection and Random Forest Algorithm (ReliefF+RF)," 2015 International Symposium on Innovations in Intelligent SysTems and Applications (INISTA), 2015, pp. 1-8.
[21] J. C. Platt, "Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines," Advances in Kernel Methods-Support Vector Learning, vol. 208, Jul. 1998.
[22] J. Malo and Camps-Valls, Gustau. (2011). "A Review of Kernel Methods in Remote Sensing Data Analysis," in Optical Remote Sensing, Springer, 2011, pp. 171–206.
[23] B. Üstün, W. J. Melssen, and L. M. C. Buydens,"Facilitating the Application of Support Vector Regression by Using a Universal Pearson VII Function Based Kernel," Chemometrics and Intelligent Laboratory Systems, vol. 81, pp 29-40, 2006.
[24] N. S. Altman, "An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression," American Statistician, vol. 46, no. 3, pp. 175-185, Aug. 1992.
[25] A. McCallum and K. Nigam, "A Comparison of Event Models for Naive Bayes Text Classification", Work Learn Text Categ, vol. 752, May 2001.
[26] L. Breiman, "Random Forests," Machine Learning, vol. 45, pp. 5-32, Oct. 2001.
[27] STMicroelectronics, "Arm®-based Cortex®-M7-core-based STM32F746," Accessed: Jun. 8, 2021. [Online]. Available: https://www.st.com/resource/en/datasheet/stm32f746vg.pdf
[28] Espressif Systems, "ESP8266," Accessed: Jun. 8, 2021. [Online]. Available: https://www.espressif.com/en/products/socs/esp8266
[29] ALLEGRO A4988 Datasheet, "A4988," Accessed: Jun. 8, 2021. [Online]. Available: https://www.allegromicro.com/-/media/files/datasheets/a4988-datasheet.ashx
[30] Nordic Semiconductor, "nRF52832," Accessed: Jun. 8, 2021. [Online]. Available: https://www.nordicsemi.com/Products/Low-power-short-range-wireless/nRF52832
[31] InvenSense, “ICM-20649,” Accessed: Jun. 8, 2021. [Online]. Available: https://product.tdk.com/en/search/sensor/mortion-inertial/imu/info?part_no=ICM-20649
[32] STMicroelectronics, “H3LIS331DL,” Accessed: April 14, 2020 [Online]. Available: https://www.st.com/resource/en/datasheet/h3lis331dl.pdf
[33] Chinese Hope Badminton Association, "National Badminton Rating Program (NBRP)," Accessed: Jun. 18, 2021. [Online]. Available: http://www.badmintonnote.com.tw/Home/grades
[34] P. Blank, J. Hossbach, D. Schuldhaus, and B. M. Eskofier, “Sensor-based Stroke Detection and Stroke Type Classification in Table Tennis,” 19th International Symposium on Wearable Computers (ISWC), 2015, pp. 93-100.
[35] C. Ó Conaire et al., "Combining Inertial and Visual Sensing for Human Action Recognition in Tennis", 9th ACM Multimedia International Conference, 2010, pp. 51-56.
[36] J. Lin et al., "Design and Implement a Mobile Badminton Stroke Classification System," 2017 19th Asia-Pacific Network Operations and Management Symposium (APNOMS), 2017, pp. 235-238.
[37] A. Anand, M. Sharma, R. Srivastava, L. Kaligounder, and D. Prakash, "Wearable Motion Sensor Based Analysis of Swing Sports," 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA), 2017, pp. 261-267.
[38] Z. Wang, M. Guo, and C. Zhao, "Badminton Stroke Recognition Based on Body Sensor Networks," in IEEE Transactions on Human-Machine Systems, vol. 46, no. 5, pp. 769-775, Oct. 2016.
[39] U. Jensen, M. Ring, and B. Eskofier, "Generic Features for Biosignal Classification," Sportinformatik 2012, vol. 112, pp. 162–168.
[40] Package weka.attributeSelection, ” Class InfoGainAttributeEval,” Accessed: May 31, 2021. [Online]. Available: https://weka.sourceforge.io/doc.dev/weka/attributeSelection/InfoGainAttributeEval.html
[41] Package weka.core, ” Class SerializationHelper,” Accessed: May 31, 2021. [Online]. Available: https://weka.sourceforge.io/doc.dev/weka/core/package-summary.html