A Weight-Aware-Based Multisource Unsupervised Domain Adaptation Method for Human Motion Intention Recognition

• Published in: IEEE transactions on cybernetics Vol. 55; no. 7; pp. 3131 - 3143
• Main Authors: Liu, Xiao-Yin, Li, Guotao, Zhou, Xiao-Hu, Liang, Xu, Hou, Zeng-Guang
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.07.2025
• Subjects: Accuracy; Algorithms; Brain modeling; Classification algorithms; Exoskeletons; Feature extraction; Generalization bound; Generators; human motion intention recognition; Humans; Intention; Movement - physiology; multisource unsupervised domain adaptation; Optimization; Pattern Recognition, Automated - methods; Robot sensing systems; Target recognition; Training; Unsupervised Machine Learning

Accurate recognition of human motion intention (HMI) is beneficial for exoskeleton robots to improve the wearing comfort level and achieve natural human-robot interaction. A classifier trained on labeled source subjects (domains) performs poorly on unlabeled target subject since the difference in in...