Multisource Information Fusion Method for Human Gait Perception

Authors

Guiliang Chen, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China
Guowei Liu, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China
Yongchao Li, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China
Chao Cai, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China
Zihao Li, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China
Dong Yang, School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401, China

Abstract

Abstract: In response to the insufficient gait perception capability during lower limb exoskeleton assistance, a human lower limb gait phase optimization classification model was proposed. A wireless transmission gait information collection system was designed for collecting the required gait phase feature information. Human joint angles were accurately calculated by fusing acceleration and angular velocity information using extended Kalman filtering. Additionally, kernel principal component analysis was applied to reduce dimensionality in conjunction with plantar pressure data. The LSSVM algorithm was employed to classify gait data, and the PSO algorithm was utilized to find the optimal classification parameters. Experimental results demonstrate that after dimensionality reduction, the PSO-based LSSVM outperforms other methods in identifying gait phases and improves timeliness, achieving an accuracy rate of 97.4%. The results validate the feasibility of using a multisource information fusion classification method for human gait phase perception, providing support for the assistance strategy of lower limb exoskeleton robots.

Recommended Citation

Chen, Guiliang; Liu, Guowei; Li, Yongchao; Cai, Chao; Li, Zihao; and Yang, Dong (2025) "Multisource Information Fusion Method for Human Gait Perception," Journal of System Simulation: Vol. 37: Iss. 10, Article 7.
DOI: 10.16182/j.issn1004731x.joss.24-0389
Available at: https://dc-china-simulation.researchcommons.org/journal/vol37/iss10/7

First Page

2522

Last Page

2532

CLC

TP391.9

Recommended Citation

Chen Guiliang, Liu Guowei, Li Yongchao, et al. Multisource Information Fusion Method for Human Gait Perception[J]. Journal of System Simulation, 2025, 37(10): 2522-2532.

Corresponding Author

Yang Dong

DOI

10.16182/j.issn1004731x.joss.24-0389