Jumping load reconstruction method based on action capture and deep learning

PDF(2330 KB)

Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (5) : 231-238.

WANG Haoqi,YU Bingqian,CHEN Jun

Author information +

History +

Abstract

Human jumping load tests with 27 participants were conducted and 9985 time histories of single-jumping load cycles and their corresponding human body motion were obtained through motion capture system, which has been commonly used in the field of medical science and biomechanics. The measured jumping cycles were expressed by a 3-parameter model and a Fourier-series model with different model parameters. The human body motion and the load parameters are considered as the inputs and the outputs of a convolutional neural network, respectively. With the trained network, the time histories of the jumping cycles could be reconstructed from the human body motion. Comparison between the reconstructed and measured jumping loads shows that the features of the reconstructed ones coincide well with those of the measured ones in both the time and the frequency domain. The proposed method can serve as an indirect technique for human jumping load measurement.

Key words

Jumping load / Motion capture / Convolutional neural network / Load reconstruction / Vibration serviceability

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

WANG Haoqi,YU Bingqian,CHEN Jun. Jumping load reconstruction method based on action capture and deep learning[J]. Journal of Vibration and Shock, 2024, 43(5): 231-238

References

[1] 陈隽. 人致荷载与人致结构振动[M]. 北京: 科学出版社, 2016 Chen J. Human-induced load and structural vibration [M]. Beijing: China Science Publishing & Media Ltd, 2016 [2] Racic V, Pavic A. Mathematical model to generate near-periodic human jumping force signals [J]. Mechanical Systems and Signal Processing, 2010, 24(1): 138-152. [3] Erlingsson S, Bodare A. Live load induced vibrations in Ullevi Stadium dynamic soil analysis [J]. Soil Dynamics and Earthquake Engineering, 1996, 15(3): 171-18 [4] Lee S, Lee K, Woo S, et al. Global vertical mode vibrations due to human group rhythmic movement in a 39 story building structure [J]. Engineering Structures, 2013, 57: 296-305 [5] http://mini.eastday.com/a/181102215430354.html [6] 谢伟平, 何卫, 艾康伟. 车站结构人行荷载特性研究[J]. 工程力学, 2012, 29(12): 256-264 Xie W, He W, Ai K. Study on characteristics of pedestrian loads of railway station structures [J]. Engineering Mechanics, 2012, 29(12): 256-264. [7] Wang, H., Chen, J., Brownjohn, J. M. W. Parameter identification of pedestrian’s spring mass damper model by ground reaction force records through a particle filter approach[J]. Journal of Sound and Vibration, 2017, 411: 409–421. [8] Wang, H., Chen, J., Nagayama, T. Parameter identification of spring-mass-damper model for bouncing people[J]. Journal of Sound and Vibration, 2019, 456: 13–29. [9] 陈隽. 人致荷载研究综述[J]. 振动与冲击, 2017, 36: 1-9. Chen J. A review of human-induced loads study [J]. Journal of Vibration and Shock, 2017, 36: 1-9. [10] Fujino, Y., Pacheco, B. M., Nakamura, S., Warnitchai, P. Synchronization of human walking observed during lateral vibration of a congested pedestrian bridge[J]. Earthquake Engineering and Structural Dynamics, 1993, 22: 741–758 [11] Caprioli, A., Manzoni, S., Zappa, E. Crowd motion measurement based on image processing[C]. Proceedings of the 26th International Modal Analysis Conference, 2008. [12] Mazzoleni, P., Zappa, E. Vision-based estimation of vertical dynamic loading induced by jumping and bobbing crowds on civil structures[J]. Mechanical Systems and Signal Processing, 2013, 33: 1–12. [13] Celik, O., Dong, C. Z., Catbas, F. N. A computer vision approach for the load time history estimation of lively individuals and crowds[J]. Computers and Structures, 2018, 200: 32–52 [14] LeCun Y, Bengio Y, Hinton G. Deep learning [J]. Nature, 2015, 521(7553): 436-444. [15] He K, Zhang X, Ren S, et al. Deep residual learning for imag recognition [C] Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, New York, 2016: 770-778. [16] 陈隽, 熊杰程. 跳跃荷载的功率谱模型研究 [J].土木工程学报, 2018, 51(9): 56-65. Chen J, Xiong J. Power spectral density model for human jumping load [J]. China Civil Engineering Journal, 2018, 51(9): 56-65 [17] 张梦诗, 陈隽, 彭怡欣, 王玲. 基于动作捕捉与刚体动力学的步行荷载重构与仿真 [J].应用基础与工程科学学报, 2013, 21(5): 961-972. Zhang M, Chen J, Peng Y, Wang L. Reproduction and simulation of walking induced load via motion capture technique and inverse dynamics of rigid body models [J]. Journal of Basic Science and Engineering, 2013, 21(5): 961-972. [18] Chen J, Wang H, Wang L. Experimental investigation on single person’s jumping load model [J]. Earthquake Engineering and Engineering Vibration, 2015, 14(4): 703-714. [19] Stefanou G., Papadrakakis M. Assessment of spectral representation and Karhunen–Loève expansion methods for the simulation of Gaussian stochastic fields [J]. Computer methods in applied mechanics and engineering, 2007, 196(21): 2465-2477. [20] 李杰, 刘章军. 基于标准正交基的随机过程展开法[J]. 同济大学学报 (自然科学版), 2006, 34(10): 1279-1283. Li J, Liu Z. Expansion method of stochastic processes based on normalized orthogonal bases [J]. Journal of Tongji University (Natural Science), 2006, 34(10): 1279-1283. [21] 陈隽,王浩祺,彭怡欣.行走激励的傅里叶级数模型及其参数的实验研究[J].振动与冲击, 2014, 33(8): 11-15+28. Chen J, Wang H, Peng Y. Experimental investigation on Fourier-series model of walking load and its coefficients [J] Journal of Vibration and Shock, 2014, 33(8): 11-15+28.