A deep Kalman filter network for hand kinematics estimation using sEMG

•A Deep Kalman Filter Network (DKFN) is proposed for hand kinematics estimation using sEMG.•DKFN exploits a novel LSTM-based Kalman filter process (LSTM-KF) to enhance sequential regression of CNN features.•In LSTM-KF, parameters of Kalman filter are learned from data using LSTM modules.•A custom wr...

Full description

Saved in:
Bibliographic Details
Published in:Pattern recognition letters 2021-03, Vol.143, p.88-94
Main Authors: Bao, Tianzhe, Zhao, Yihui, Zaidi, Syed Ali Raza, Xie, Shengquan, Yang, Pengfei, Zhang, Zhiqiang
Format: Article
Language:English
Subjects:
Artificial neural networks
Computer applications
Deep learning
Electromyography
Feature extraction
Hand
Hand (anatomy)
Hand kinematics estimation
Interfaces
Kalman filter
Kalman filters
Kinematics
Learning algorithms
Long short-term memory
LSTM
Man-machine interfaces
Neural networks
Parameter estimation
sEMG
Sequential regression
Wrist
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A Deep Kalman Filter Network (DKFN) is proposed for hand kinematics estimation using sEMG.•DKFN exploits a novel LSTM-based Kalman filter process (LSTM-KF) to enhance sequential regression of CNN features.•In LSTM-KF, parameters of Kalman filter are learned from data using LSTM modules.•A custom wrist dataset and a public finger dataset are leveraged for model evaluation. In human-machine interfaces (HMI), deep learning (DL) techniques such as convolutional neural networks (CNN), long-short term memory networks (LSTM) and the hybrid CNN-LSTM framework have been exploited for hand kinematics estimation using surface electromyography (sEMG). However, these DL techniques only capture the relationship between sEMG and hand kinematics, but ignores the prior knowledge of the system. By contrast, Kalman filter (KF) can apply Kalman gain to combine the internal transition model and the observation model effectively. To this end, we propose a novel architecture named deep Kalman filter network (DKFN), in which we utilize CNN to extract high-level features from sEMG and employ a LSTM-based Kalman filter process (LSTM-KF) to conduct sequential regression. In particular, LSTM-KF adopts the computational graph of KF but estimates parameters of the transition/observation model and the Kalman gain from data using LSTM modules. With this process, the advantages of KF and LSTM can be exploited jointly. Experimental results demonstrate that the proposed DKFN can outperform CNN and CNN-LSTM in the sequential regression for wrist/fingers kinematics estimation.
ISSN:0167-8655
1872-7344
DOI:10.1016/j.patrec.2021.01.001