|
|
基于混合数据的下肢外骨骼机器人步态活动识别方法研究
|
Abstract:
下肢外骨骼机器人步态活动识别是实现外骨骼智能化与人机协同的关键技术，对康复医学、辅助行走以及运动增强等领域具有重要意义。然而，现有步态数据采集面临硬件布设复杂、成本高昂及采集过程耗时长等问题，同时，传统的步态特征手工提取方法自动化程度低，且主观性强，工作量大且繁琐。针对这些问题，本文提出了一种基于仿真与真实混合数据的步态活动识别方法。首先，利用Solidworks对下肢外骨骼机器人进行建模，并通过Simulink搭建出下肢外骨骼机器人动力学模型，以C3D人体步态数据库为数据源，生成所需的步态仿真数据。随后，将仿真数据与WISDM数据集中的真实步态活动数据进行融合，构建了出混合数据集，从而显著降低实际实验的成本和难度。随后，针对时序数据的特点，设计出一种用于下肢外骨骼机器人步态活动识别的深度学习模型——Halfconstm，以对步态活动进行精准识别与分类。实验结果表明，该方法能够高效、准确地完成步态活动识别任务，具有较高的应用潜力，为下肢外骨骼机器人的智能化与便捷性设计提供了新的解决方案。
Gait recognition of lower extremity exoskeleton robot is a key technology to realize the intelligent exoskeleton and human-machine collaboration, which is of great significance in the fields of rehabilitation medicine, assisted walking, and motion enhancement. However, the existing gait data acquisition is faced with such problems as complex hardware layout, high cost, and long collection process. At the same time, the traditional manual gait feature extraction method has a low degree of automation, strong subjectivity, and a heavy workload. To solve these problems, this paper proposes a gait recognition method based on mixed simulation and real data. First of all, Solidworks was used to model the lower limb exoskeleton robot. The dynamics model of the lower limb exoskeleton robot was built through Simulink, and the required gait simulation data was generated using the C3D human gait database as the data source. The simulation data was then fused with the real gait activity data from the WISDM dataset to construct a hybrid dataset, which significantly reduced the cost and difficulty of the actual experiment. Then, according to the characteristics of time series data, a deep learning model, Halfconstm, is designed for gait recognition of lower limb exoskeleton robots to accurately identify and classify gait movements. The experimental results show that this method can efficiently and accurately complete the task of gait recognition, which has high application potential and provides a new solution for the intelligent and convenient design of lower limb exoskeleton robots.
| [1] | 郝帅, 田卓, 马旭, 等. 混合数据驱动的轻量化YOLOv5故障选线方法[J]. 西安科技大学学报, 2024, 44(5): 966-975. |
| [2] | 刘艺, 秦伟, 李庚松, 等. 面向混合特征数据的粒子群填补方法[J]. 国防科技大学学报, 2024, 46(6): 107-112. |
| [3] | Catal, C., Tufekci, S., Pirmit, E. and Kocabag, G. (2015) On the Use of Ensemble of Classifiers for Accelerometer-Based Activity Recognition. Applied Soft Computing, 37, 1018-1022. https://doi.org/10.1016/j.asoc.2015.01.025 |
| [4] | Ermes, M., Parkka, J., Mantyjarvi, J. and Korhonen, I. (2008) Detection of Daily Activities and Sports with Wearable Sensors in Controlled and Uncontrolled Conditions. IEEE Transactions on Information Technology in Biomedicine, 12, 20-26. https://doi.org/10.1109/titb.2007.899496 |
| [5] | He, Z. and Jin, L. (2009) Activity Recognition from Acceleration Data Based on Discrete Consine Transform and SVM. 2009 IEEE International Conference on Systems, Man and Cybernetics, San Antonio, 11-14 October 2009, 5041-5044. https://doi.org/10.1109/icsmc.2009.5346042 |
| [6] | Jain, A. and Kanhangad, V. (2018) Human Activity Classification in Smartphones Using Accelerometer and Gyroscope Sensors. IEEE Sensors Journal, 18, 1169-1177. https://doi.org/10.1109/jsen.2017.2782492 |
| [7] | Han, J., Lee, G.H., Park, S., Lee, J. and Choi, J.K. (2022) A Multivariate-Time-Series-Prediction-Based Adaptive Data Transmission Period Control Algorithm for IoT Networks. IEEE Internet of Things Journal, 9, 419-436. https://doi.org/10.1109/jiot.2021.3124673 |
| [8] | Liu, C., Zhen, J. and Shan, W. (2023) Time Series Classification Based on Convolutional Network with a Gated Linear Units Kernel. Engineering Applications of Artificial Intelligence, 123, Article ID: 106296. https://doi.org/10.1016/j.engappai.2023.106296 |
| [9] | Liu, F., Zhang, Z. and Zhou, R. (2022) Automatic Modulation Recognition Based on CNN and GRU. Tsinghua Science and Technology, 27, 422-431. https://doi.org/10.26599/tst.2020.9010057 |
| [10] | Zhang, M., Li, W., Tao, R., Li, H. and Du, Q. (2022) Information Fusion for Classification of Hyperspectral and Lidar Data Using IP-CNN. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-12. https://doi.org/10.1109/tgrs.2021.3093334 |
| [11] | Weis,s G. (2019) WISDM Smartphone and Smartwatch Activity and Biometrics Dataset Data Set. UCI Machine Learning Repository. https://archive.ics.uci.edu/ml/datasets/WISDM+Smartphone+and+Smartwatch+Activity+and+Biometrics+Dataset |
| [12] | Münzner, S., Schmidt, P., Reiss, A., Hanselmann, M., Stiefelhagen, R. and Dürichen, R. (2017) CNN-Based Sensor Fusion Techniques for Multimodal Human Activity Recognition. Proceedings of the 2017 ACM International Symposium on Wearable Computers, Maui, 11-15 September 2017, 158-165. https://doi.org/10.1145/3123021.3123046 |
| [13] | Bevilacqua, A., MacDonald, K., Rangarej, A., Widjaya, V., Caulfield, B. and Kechadi, T. (2019) Human Activity Recognition with Convolutional Neural Networks. In: Brefeld, U., et al., Eds., Machine Learning and Knowledge Discovery in Databases, Springer, 541-552. https://doi.org/10.1007/978-3-030-10997-4_33 |
| [14] | Rahman, M.A., Mia, Y., Masum, M.R., Abid, D.M.H. and Islam, T. (2022) Real Time Human Activity Recognition from Accelerometer Data Using Convolutional Neural Networks. 2022 7th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, 22-24 June 2022, 1394-1397. https://doi.org/10.1109/icces54183.2022.9835797 |
| [15] | Cooley, J. and Tukey, J. (1965) An Algorithm for Machine Calculation of Complex Fourier Series. IEEE Press, 223-227. |
| [16] | Ronao, C.A. and Cho, S. (2016) Human Activity Recognition with Smartphone Sensors Using Deep Learning Neural Networks. Expert Systems with Applications, 59, 235-244. https://doi.org/10.1016/j.eswa.2016.04.032 |
| [17] | Gholamrezaii, M. and AlModarresi, S. (2021) A Time-Efficient Convolutional Neural Network Model in Human Activity Recognition. Multimedia Tools and Applications, 80, 19361-19376. https://doi.org/10.1007/s11042-020-10435-1 |
| [18] | Chen, J., Kao, S., He, H., Zhuo, W., Wen, S., Lee, C., et al. (2023) Run, Don’t Walk: Chasing Higher FLOPS for Faster Neural Networks. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, 17-24 June 2023, 12021-12031. https://doi.org/10.1109/cvpr52729.2023.01157 |
