Can data processing algorithms ensure sufficient accuracy to estimate human body pose via wearable systems with use of IMU sensors? – an experimental evaluation

• Autorzy: Szczerba Aleksandra , Prochor Piotr , Piszczatowski Szczepan

Identyfikatory

DOI

10.24425/bpasts.2024.148835

• Języki publikacji: EN

Abstrakty

EN

Background: The aim of the study was to answer two questions: 1 – Can data processing algorithms ensure sufficient accuracy for estimating human body pose via wearable systems? 2 – How to process the IMU sensor data to obtain the most accurate information on the human body pose? To answer these questions, the authors evaluated proposed algorithms in terms of accuracy and reliability. Methodology: data acquisition was performed with tested IMU sensors system mounted onto a Biodex System device. Research included pendulum movement with seven angular velocities (10-120°/s) in five angular movement ranges (30-120°). Algorithms used data from accelerometers and gyroscopes and considered complementary and/or Kalman filters with adjusted parameters. Moreover, angular velocity registration quality was also taken into consideration. Results: differences between means for angular velocity were 0.55÷1.05°/s and 1.76÷3.11%. In the case of angular position relative error of means was 4.77÷10.84%, relative error of extreme values was 2.15÷4.81% and Spearman’s correlation coefficient was 0.74÷0.89. Conclusions: Algorithm calculating angles based on acceleration-derived quaternions and with implementation of Kalman filter was the most accurate for data processing and can be adapted for future work with IMU sensors systems, especially in wearable devices that are designated to support human in daily activity.

Słowa kluczowe

EN

pose estimation; wearable electronics; motion sensors; inertial measurement unit; human motion tracking

PL

estymacja pozycji; elektronika noszona; czujnik ruchu; inercyjna jednostka pomiarowa; pomiar ruchu człowieka

• Wydawca: Polska Akademia Nauk, Wydział IV Nauk Technicznych
• Czasopismo: Bulletin of the Polish Academy of Sciences. Technical Sciences
• Rocznik: 2024
• Tom: Vol. 72, nr 1
• Strony: art. no. e148835
• Opis fizyczny: Bibliogr. 39 poz., rys., tab.

Twórcy

autor

Szczerba Aleksandra

• Department of Biomaterials and Medical Devices Engineering, Institute of Biomedical Engineering, Faculty of Mechanical Engineering, BialystokUniversity of Technology, Wiejska 45C Street, 15-351 Bialystok, Poland

• https://orcid.org/0000-0003-1371-5489

autor

Prochor Piotr

• Department of Biomaterials and Medical Devices Engineering, Institute of Biomedical Engineering, Faculty of Mechanical Engineering, BialystokUniversity of Technology, Wiejska 45C Street, 15-351 Bialystok, Poland

• https://orcid.org/0000-0001-9988-3859

autor

Piszczatowski Szczepan

• Department of Biomaterials and Medical Devices Engineering, Institute of Biomedical Engineering, Faculty of Mechanical Engineering, BialystokUniversity of Technology, Wiejska 45C Street, 15-351 Bialystok, Poland

• https://orcid.org/0000-0003-2191-3979

Bibliografia

• [1] K. Lee and W.A. Tang, “A Fully Wireless Wearable Motion Tracking System with 3D Human Model for Gait Analysis,” Sensors, vol. 21, no. 12, p. 4051, 2021, doi: 10.3390/s21124051.
• [2] A. Blanco Ortega et al. “Biomechanics of the Upper Limbs: A Review in the Sports Combat Ambit Highlighting Wearable Sensors,” Sensors, vol. 22, no. 13, p. 4905, 2022, doi: 10.3390/s22134905.
• [3] Y. Jiang, P. Malliaras, B. Chen, and D. Kulić, “Real-time forecasting of exercise-induced fatigue from wearable sensors,” Comput. Biol. Med., vol. 148, p. 105905, 2022, doi: 10.1016/j.compbiomed.2022.105905.
• [4] R. Argent, P. Slevin, A. Bevilacqua, M. Neligan, A. Daly, and B. Caulfield, “Clinician perceptions of a prototype wearable exercise biofeedback system for orthopaedic rehabilitation: a qualitative exploration,” BMJ Open, vol. 8, no. 10, p. e026326, 2018, doi: 10.1136/bmjopen-2018-026326.
• [5] M. Ollenschläger et al., “Wearable gait analysis systems: ready to be used by medical practitioners in geriatric wards?,” Eur. Geriatr. Med., vol. 13, no. 4, pp. 817–824, 2022, doi: 10.1007/s41999-022-00629-1.
• [6] I.M. Pires, F. Hussain, G. Marques, and N.M. Garcia, “Comparison of machine learning techniques for the identification of human activities from inertial sensors available in a mobile device after the application of data imputation techniques,” Comput. Biol. Med., vol. 135, p. 104638, 2021, doi: 10.1016/j.compbiomed.2021.104638.
• [7] Y. Guo et al., “High-accuracy wearable detection of freezing of gait in Parkinson’s disease based on pseudo-multimodal features,” Comput. Biol. Med., vol. 146, p. 105629, 2022, doi: 10.1016/j.compbiomed.2022.105629.
• [8] W. Xu, Y. Zheng, C.H. Chu, L. Cheng, and J. Kim, “Applying deep learning technology for automatic fall detection using mobile sensors,” Biomed. Signal. Proces., vol. 72 (part B), p. 103355, 2022, doi: 10.1016/j.bspc.2021.103355.
• [9] F. Ryser, S. Hanassab, O. Lambercy, E. Werth, and R. Gassert, “Respiratory analysis during sleep using a chest-worn accelerometer: A machine learning approach,” Biomed. Signal. Proces., vol. 78, pp. 104014, 2022, doi: 10.1016/j.bspc.2022.104014.
• [10] L.M. Dang, K. Min, H. Wang, M.J. Piran, C.H. Lee, and H. Moon, “Sensor-based and vision-based human activity recognition: A comprehensive survey,” Pattern Recogn., vol. 108, p. 107561, 2020, doi: 10.1016/j.patcog.2020.107561.
• [11] Z. Zhang, Z. Wang, H. Lei, and W. Gu, “Gait phase recognition of lower limb exoskeleton system based on the integrated network model,” Biomed. Signal. Proces., vol. 76, p. 103693, 2022, doi: 10.1016/j.bspc.2022.103693.
• [12] V.T. Van Hees et al. “Separating movement and gravity components in an acceleration signal and implications for the assessment of human daily physical activity,” Plos One, vol. 8, no. 4, pp. e61691, 2013, doi: 10.1371/journal.pone.0061691.
• [13] K.R. Vidyarani, V. Talasila, N. Megharjun, M. Supriya, K.R. Prasad, and G.R. Prashanth, “An inertial sensing mechanism for measuring gait parameters and gait energy expenditure,” Biomed. Signal. Proces., vol. 70, p. 103056, 2021, doi: 10.1016/j.bspc.2021.103056.
• [14] Z. Zheng, Q. Wang, D. Deng, Q. Wang, and W. Huang, “CG-Recognizer: A biosignal-based continuous gesture recognition system,” Biomed. Signal. Proces., vol. 78, p. 103995, 2022, doi: 10.1016/j.bspc.2022.103995.
• [15] S. Lambrecht, S.L. Nogueira, M. Bortole, A.A.G. Siqueira, M.H. Terra, E. Rocon, and J.L. Pons, “Inertial sensor error reduction through calibration and sensor fusion,” Sensors, vol. 16, no. 2, p. 235, 2016, doi: 10.3390/s16020235.
• [16] J.F.S. Lin and D. Kulić, “Human pose recovery using wireless inertial measurement units,” Physiol. Meas., vol. 33, no. 12, pp. 2099–2115, 2012, doi: 10.1088/0967-3334/33/12/2099.
• [17] T.L. Baldi, F. Farina, A. Garulli, A. Giannitrapani, and D. Prattichizzo, “Upper Body Pose Estimation Using Wearable Inertial Sensors and Multiplicative Kalman Filter,” IEEE Sens., vol. 20, no. 1, pp. 492–500, 2019, doi: 10.1109/JSEN.2019.2940612.
• [18] J.F.B. Ruiz, J.D. Chaparro, C.B. Peño, H.A.L. Solano, X.T. García and J.C. López, “A low-cost and unobtrusive system for fall detection,” Procedia Comput. Sci., vol. 192, pp. 2160–2169, 2021, doi: 10.1016/j.procs.2021.08.229.
• [19] Y. Kim, H. Jung, B. Koo, J. Kim, T. Kim, and Y. Nam, “Detection of Pre-Impact Falls from Heights Using an Inertial Measurement Unit Sensor,” Sensors, vol. 20, no. 18, p. 5388, 2020, doi: 10.3390/s20185388.
• [20] T. Amasay, K. Zodrow, L. Kincl, J. Hess, and A. Karduna, “Validation of tri-axial accelerometer for the calculation of elevation angles,” Int. J. Ind. Ergonom., vol. 39, no. 5, pp. 783–789, 2009, doi: 10.1016/j.ergon.2009.03.005.
• [21] K. Feng et al. “A new quaternion-based Kalman filter for real-time attitude estimation using the two-step geometrically-intuitive correction algorithm,” Sensors, vol. 17, no. 9, p. 2146, 2017, doi: 10.3390/s17092146.
• [22] W.H.K. De Vries, H.E.J. Veeger, C.T.M. Baten, and F.C.T. van der Helm, “Magnetic distortion in motion labs, implications for validating inertial magnetic sensors,” Gait Posture, vol. 29, no. 4, pp. 535–541, 2009, doi: 10.1016/j.gaitpost.2008.12.004.
• [23] N.H. Ariffin, N. Arsad, and B. Bais, “ Low cost MEMS gyroscope and accelerometer implementation without Kalman Filter for angle estimation,” in Proc. 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering. (ICAEES), 2016, pp. 77–82, doi: 10.1109/icaees.2016.7888013.
• [24] J. Abbasi, H. Salarieh, and A. Alasty, “A motion capture algorithm based on inertia-Kinect sensors for lower body elements and step length estimation,” Biomed. Signal. Proces., vol. 64, p. 102290, 2021, doi: 10.1016/j.bspc.2020.102290.
• [25] L. Xue, C.Y. Jiang, H.L. Chang, Y. Yang, W. Qin and W.Z. Yuan, “A novel Kalman filter for combining outputs of MEMS gyroscope array,” Measurement, vol. 45, no. 4, pp. 745–754, 2012, doi: 10.1016/j.measurement.2011.12.016.
• [26] P. Gui, L. Tang and S. Mukhopadhyay, “MEMS based IMU for tilting measurement: Comparison of complementary and Kalman filter based data fusion,” in Proc. 2015 IEEE 10th conference on Industrial Electronics and Applications. (ICIEA), 2015, pp. 2004–2009, doi: 10.1109/iciea.2015.7334442.
• [27] C. Yi, J. Ma, H. Guo, J. Han, H. Gao, F. Jiang, and C. Yang, “Estimating three-dimensional body orientation based on an improved complementary filter for human motion tracking,” Sensors, vol. 18, no. 11, p. 3765, 2018, doi: 10.3390/s18113765.
• [28] G. Ligorio and A.M. Sabatini, “A novel Kalman filter for human motion tracking with an inertial-based dynamic inclinometer,” IEEE Trans. Biomed. Eng., vol. 62, no. 8, pp. 2033–2043, 2015, doi: 10.1109/TBME.2015.2411431.
• [29] P. Promrit, S. Chokchaitam, and M. Ikura, “ In-vehicle MEMS IMU calibration using accelerometer,” in Proc. 2018 IEEE 5th International Conference on Smart Instrumentation, Measurement and Application. (ICSIMA), 2018, pp. 1–3, doi: 10.1109/icsima.2018.8688778.
• [30] R.G. Valenti, I. Dryanovski, and J. Xiao, “Keeping a good attitude: A quaternion-based orientation filter for IMUs and MARGs,” Sensors, vol. 15, no. 8, pp. 19302–19330, 2015, doi: 10.3390/s150819302.
• [31] R.G. Valenti, I. Dryanovski, and J. Xiao, “A linear Kalman filter for MARG orientation estimation using the algebraic quaternion algorithm,” IEEE T. Instrum. Meas., vol. 65, no. 2, pp. 467–481, 2015, doi: 10.1109/TIM.2015.2498998.
• [32] S.O.H. Madgwick, “ AHRS algorithms and calibration solutions to facilitate new applications using low-cost MEMS,” PhD Thesis, University of Calgary, Canada, 2000.
• [33] D.K. Shaeffer, “MEMS inertial sensors: A tutorial overview,” IEEE Commun. Mag., vol. 51, no. 4, pp. 100–109, 2013, doi: 10.1109/MCOM.2013.6495768.
• [34] H.W. Hsu, T.Y. Wu, S. Wan, W.H. Wong and C.Y. Lee, “QuatNet: Quaternion-Based Head Pose Estimation With Multiregression Loss,” IEEE Trans. Multimedia, vol. 21, no. 4, pp. 1035–1046, 2018, doi: 10.1109/TMM.2018.2866770.
• [35] H. Guo and H. Hong, “Research on filtering algorithm of MEMS gyroscope based on information fusion,” Sensors, vol. 19, no. 16, pp. 3552, 2019, doi: 10.3390/s19163552.
• [36] C. Kownacki, “Optimization approach to adapt Kalman filters for the real-time application of accelerometer and gyroscope signals’ filtering,” Digit. Signal. Process., vol. 21, no. 1, pp. 131–140, 2011, doi: 10.1016/j.dsp.2010.09.001.
• [37] P. Schober, C. Boer and L.A. Schwarte, “Correlation coefficients: appropriate use and interpretation,” Anesth. Analg., vol. 126, no. 5, pp. 1763–1768, 2018, doi: 10.1213/ANE.0000000000002864.
• [38] N. Jalloul, “Wearable sensors for the monitoring of movement disorders,” Biomed. J., vol. 41, no 4, pp. 249–253, 2018, doi: 10.1016/j.bj.2018.06.003.
• [39] F.J. Ordóñez and D. Roggen, “Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition,” Sensors, vol. 16, no. 1, p. 115, 2016, doi: 10.3390/s16010115.

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2024).