在工廠量產產品時，依據產品規格與產量指派產線作業員進行產品裝配，但由於產線員工裝配產品時，易發生人為操作失誤導致產品破裂或是產品漏裝，進而導致影響整體產品出貨品質與製造成本。
故本篇論文提出以作業員戴上指套且指套上黏附九軸感測器和手握自動螺絲起子，蒐集動作感測數據，再以卡爾曼濾波進行感測數據濾波，以越零率方式進行動作切割，再利用Mahony 濾波計算誤差模型來校正感測器數據後計算出四元數，利用四元數當作特徵後利用一維卷積神經網路來分類 (1) 右上頂螺絲、(2) 右下頂螺絲、(3) 右斜上頂螺絲、(4) 左斜下頂螺絲、(5) 向下鎖螺絲、(6) 抬手放下，共六種作業員裝配基本動作，由此六種裝配動作組合順序以判別裝配產品是否正確。
實驗結果發現利用Mahony 濾波校正感測器後再計算出四元數為特徵進行一維卷積神經網路1DCNN 能提升72% 的辨識率，進行長短期記憶模型LSTM 能提升30%辨識率，進行CNN-LSTM 也能提升70%的辨識率，而校正後的數據在1DCNN 有高達94%的平均辨識率，比使用LSTM 模型高39%辨識率，比使用CNN-LSTM 高3%辨識率，故使用此方法能預防現今作業員鎖螺絲動作出錯與降低作業員動作訓練成本，與改善產線員工管理智能化不足等問題。

When mass-producing products in the factory, the production line operators are assigned to carry out product assembly based on product specifications and product number. However, when the production line employees assemble products, human operation errors are prone to cause product rupture or product miss screw , which in turn affects the overall product quality and manufacturing cost.
Therefore, This paper proposes that the operator wears a finger sleeve with a nine-axis sensor attached and holds an automatic screwdriver to collect motion sensing data, and then uses the Kalman filter to filter the sensing data, and performs motion cutting with a zerocrossing rate. Then use Mahony filter to calculate the error model to correct the sensor data and calculate the quaternion, use the quaternion as a feature, and then use a one-dimensional Convolutional neural network to classify (1) upper right screw , (2) lower right screw ,(3) right diagonal screw,(4) Left diagonal screw,(5) down lock screw ,(6) Raise hand and lay down , total of six basic assembly actions for operators, The sequence of six assembly actions is used to judge whether the assembled product is correct.
The experimental results found that the use of Mahony filter to calibrate the sensor and then calculate the quaternion as the feature for one-dimensional convolutional neural network 1DCNN can increase the recognition rate by 72%, and the long- and short-term memory model LSTM can increase the recognition rate by 30%. CNN-LSTM can also increase the recognition rate by 70%, and the corrected data has an average recognition rate of 94% in 1DCNN, which is 39% higher than the LSTM model and 3% higher than the CNN-LSTM.Using this method can prevent the current operator from locking screws and reduce the cost of operator training, and improve the lack of intelligent management of production line employees.

[1] Bastian C. Müllera,The Duy Nguyenb, Quang-Vinh Dangc, Bui Minh Ducc, Günther Seligera ,Jörg Krügerb, Holger Kohld, “Motion tracking applied in assembly for worker training in different locations”, The 23rd CIRP Conference on Life Cycle Engineering, Volume 48, pp. 460-465,2016
[2] Gartner Says Global End-User Spending on Wearable Devices to Total $52 Billion in2020,2019[Online].Available:https://www.gartner.com/en/newsroom/press-releases/2019-10-30-gartner-says-global-end-user-spending-on-wearable-dev
[3] Yi-Chen Huang , Tsung-Long Chen , Bo-Chun Chiu ,Chih-Wei Yi , Chung-Wei Lin ; Yu-Jung Yeh ; Lun-Chia Kuo, "Calculate Golf Swing Trajectories from IMU Sensing Data," 2012 41st International Conference on Parallel Processing Workshops, Pittsburgh, PA, pp. 505-513,2012
[4] Anguita, Davide,Ghio, Alessandro,Oneto, Luca,Parra Perez, Xavier. “A public domain dataset for human activity recognition using smartphones. “,A: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. "Proceedings of the 21th International European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning". , pp. 437-442,2013
[5] Y. Guo, L. Yang, X. Ding, J. Han and Y. Liu, "OpenSesame: Unlocking smart phone through handshaking biometrics," 2013 Proceedings IEEE INFOCOM, Turin, pp. 365-369,2013
[6] F.Pilatia, Maurizio Facciob, Mauro Gamberic, Alberto Regattieri, “Learning manual assembly through real-time motion capture for operator training with augmented reality”, 10th Conference on Learning Factories,pp189-195,2020
[7] C. Munroe, Y. Meng, H. Yanco and M. Begum, "Augmented reality eyeglasses for promoting home-based rehabilitation for children with cerebral palsy," 2016 11th ACM/IEEE International Conference on Human-Robot Interaction (HRI), Christchurch, pp. 565-565,2016
[8] M.T.D. G. N. M. Karunarathna, C.S. A. Siriwardana and M. Y. R. Dharmawardana, "An Activity Analysis to Investigate the Root Causes of Worker Productivity Losses in Sri Lankan Building Construction Project," 2019 Moratuwa Engineering Research Conference (MERCon), Moratuwa, Sri Lanka, pp. 412-417,2019
[9] Y.Seol Lee,Sung-Bae Cho, "Activity Recognition Using Hierarchical Hidden Markov Models on a Smartphone with 3D Accelerometer " Hybrid Artificial Intelligent Systems: 6th International Conference, pp.460-467,2011
[10] S.Dara and P. Tumma,"Feature Extraction By Using Deep Learning: A Survey," 2018 Second International Conference on Electronics, Communication and Aerospace Technology (ICECA), Coimbatore, pp. 1795-1801,2018
[11] F.J. Ordóñez, D. Roggen, “Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition”. Sensors 2016, vol 16, pp. 115,2016
[12] N.Michel, "Recurrent neural networks: overview and perspectives," Proceedings of the 2003 International Symposium on Circuits and Systems, pp. III-III, 2003
[13] S. Albawi, T. A. Mohammed and S. Al-Zawi, "Understanding of a convolutional neural network," 2017 International Conference on Engineering and Technology (ICET), pp. 1-6,2017
[14] Drumond, Rafael Rego, Bruno A. D. Marques, Cristina Nader Vasconcelos and Esteban Walter Gonzalez Clua. “PEEK - An LSTM Recurrent Network for Motion Classification from Sparse Data.” VISIGRAP,2018
[15] S. Albawi, T. A. Mohammed and S. Al-Zawi, "Understanding of a convolutional neural network," 2017 International Conference on Engineering and Technology, pp. 1-6,2017
[16] R. E. Kalman, A new approach to linear filtering and prediction problems, Journal of basic Engineering, Volume 82 , no.1 , pp35-45,1960
[17] mblientlab, Metawearc product specification v1.0, 2018 [Online]. Available: https://mbientlab.com/documents/MetaWearC-CPRO-PS.pdf
[18] Keras, Keras API ,2020 [Online]. Available: https://keras.io/api/
[19] D. Tedaldi, A. Pretto and E. Menegatti, "A robust and easy to implement method for IMU calibration without external equipments," 2014 IEEE International Conference on Robotics and Automation (ICRA), pp. 3042-3049,2014
[20] H. El-Ghaish, M. E. Hussien, A. Shoukry and R. Onai, "Human Action Recognition Based on Integrating Body Pose, Part Shape, and Motion," in IEEE Access, vol. 6, pp. 49040-49055, 2018
[21] P. Elias, J. Sedmidubsky and P. Zezula, "Understanding the Gap between 2D and 3D Skeleton-Based Action Recognition," 2019 IEEE International Symposium on Multimedia (ISM), pp. 192-1923,2019
[22] Huang, S. , Tang, J. , Dai, J. , Wang, Y. Signal Status Recognition Based on 1DCNN and Its Feature Extraction Mechanism Analysis. Sensors ,pp19,2019
[23] F. Hernández, L. F. Suárez, J. Villamizar and M. Altuve, "Human Activity Recognition on Smartphones Using a Bidirectional LSTM Network," 2019 XXII Symposium on Image, Signal Processing and Artificial Vision (STSIVA), Bucaramanga, Colombia, pp. 1-5,2019,