A Multimodal Fusion Model for Estimating Human Hand Force Comparing surface electromyography and ultrasound signals

doi:10.1109/MRA.2022.3177486

CORC > 自动化研究所 > 中国科学院自动化研究所 > 复杂系统管理与控制国家重点实验室 > 先进机器人控制团队

	A Multimodal Fusion Model for Estimating Human Hand Force Comparing surface electromyography and ultrasound signals
	Zou, Yongxiang 1,2; Cheng, Long 1,2; Li, Zhengwei 1
刊名	IEEE ROBOTICS & AUTOMATION MAGAZINE
	2022-12-01
卷号	29 期号:4 页码:10-24
关键词	Force Ultrasonic imaging Muscles Robots Feature extraction Estimation Exoskeletons
ISSN号	1070-9932
DOI	10.1109/MRA.2022.3177486
通讯作者	Zou, Yongxiang(zouyongxiang2019@ia.ac.cn)
英文摘要	Biomimetic robots have received significant attention in recent years. Among them, the wearable exoskeleton, which imitates the functions of the musculoskeletal system to assist humans, is a typical biomimetic robot. Given that safe human-robot interaction plays a critical role in the successful application of wearable exoskeletons, this work studies the clinical readiness of a multimodal fusion model that estimates hand force based on the surface electromyography (sEMG) and A-mode ultrasound signals of the forearm muscles. The proposed multimodal fusion model affords the biomimetic hand exoskeleton assisting the elderly in completing daily tasks or quantitatively assessing the recovery level of poststroke patients. The suggested fusion model is called Optimization of Latent Representation for the Self-Attention Convolutional Neural Network (OLR-SACNN), which utilizes a common component extraction module (CCEM) and a complementary component retention module (CCRM) to optimize latent representation of the multiple modalities. Then the optimized latent representations are fused with the self-attention mechanism. The experiments conducted on a self-collected multimodal data set verify performance of the proposed OLR-SACNN model. Specifically, compared to solely employing sEMG or A-mode ultrasound signals, the force estimation's normalized mean-square error (NMSE) based on the multiple modalities decreases by 97.7 and 38.92%, respectively. Furthermore, the OLR-SACNN model has been used to estimate the hand force of some poststroke patients and attained the desired performance.
资助项目	National Natural Science Foundation of China[62025307] ; National Natural Science Foundation of China[U1913209] ; Beijing Municipal Natural Science Foundation[JQ19020] ; Department of Mathematics and Theories, Peng Cheng Laboratory, China
WOS关键词	EXOSKELETON
WOS研究方向	Automation & Control Systems ; Robotics
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000900084900004
资助机构	National Natural Science Foundation of China ; Beijing Municipal Natural Science Foundation ; Department of Mathematics and Theories, Peng Cheng Laboratory, China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/51105]
专题	自动化研究所_复杂系统管理与控制国家重点实验室_先进机器人控制团队
通讯作者	Zou, Yongxiang
作者单位	1.Inst Automation, Chinese Acad Sci, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zou, Yongxiang,Cheng, Long,Li, Zhengwei. A Multimodal Fusion Model for Estimating Human Hand Force Comparing surface electromyography and ultrasound signals[J]. IEEE ROBOTICS & AUTOMATION MAGAZINE,2022,29(4):10-24.
APA	Zou, Yongxiang,Cheng, Long,&Li, Zhengwei.(2022).A Multimodal Fusion Model for Estimating Human Hand Force Comparing surface electromyography and ultrasound signals.IEEE ROBOTICS & AUTOMATION MAGAZINE,29(4),10-24.
MLA	Zou, Yongxiang,et al."A Multimodal Fusion Model for Estimating Human Hand Force Comparing surface electromyography and ultrasound signals".IEEE ROBOTICS & AUTOMATION MAGAZINE 29.4(2022):10-24.