Estimate the Kinematics with EMG Signal Using Fuzzy Wavelet Neural Network for Biomechanical Leg Application
Several linear and nonlinear models were proposed to predict the forward relationship between EMG signals and kinematics for biomechanical limbs, which is meaningful for EMG-based control. Although using nonlinear model to predict the kinematics is able t
- PDF / 506,040 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 36 Downloads / 192 Views
School of Mechanical Engineering, Northwestern Polytechnical University, Youyi xilu 127 hao, Xi’an, People’s Republic of China [email protected] 2 Images, Signals and Intelligence Systems Laboratory (LISSI/EA 3956), UPEC, Senart-Fontainebleau Institute of Technology, Bât.A, 77127 Lieusaint, France
Abstract. Several linear and nonlinear models were proposed to predict the forward relationship between EMG signals and kinematics for biomechanical limbs, which is meaningful for EMG-based control. Although using nonlinear model to predict the kinematics is able to represent rational complex relationship between EMG signals and desired outputs, there exists high risk for overfitting models to training data and calculating burden because of the multi-channel variation EMG signals. Inspired by the hypothesis that CNS modulates muscle synergies to simplify the motor control and learning of coordinating variation of redundant joints, this paper proposed to extract the synergies to reduce the dimension of EMG-based control. Furthermore, the fuzzy wavelet neural network was developed to generate velocity–adapted gait by the reference gaits only with the limited set of experimental trials. The experimental results show the efficiency and robust of this approach. Keywords: EMG
Muscle synergy Fuzzy wavelet neural network
1 Introduction The myoelectric signal is the electrical manifestation of muscular contractions, which reflexes the plentiful neural control information [1]. EMG based control has advantages for achieving intuitive simultaneous and proportional control of exoskeletons, myoelectric prostheses and bio-robot, and moreover for the development of novel diagnostic tools and rehabilitation approaches. In past few years, several contributions were proposed to predict the forward relationship between EMG signals and kinematics for biomechanical limb. Such as the approaches based on the linear models. A linear ‘mixing matrix’ was built to provides an explicit expression of the EMG in terms of force functions, which correspond to the wrist intended activations of physiological DOFs of natural movement [2]. In their later study, muscle synergies of both DOF were extracted at once, and three synergies can achieve simultaneous control when applying separately on each DOF [3]. © Springer International Publishing Switzerland 2016 Y. Tan et al. (Eds.): ICSI 2016, Part II, LNCS 9713, pp. 132–140, 2016. DOI: 10.1007/978-3-319-41009-8_14
Estimate the Kinematics with EMG Signal
133
Nonlinear model based approach are not as reliant on robust features as linear models, and therefore are able to represent rational complex relationship between synergies and desired outputs. J.M. Hahne systematically compare linear and nonlinear regression techniques including linear regression, mixture of linear experts, multilayerperceptron and kernel ridge regression, for an independent, simultaneous and proportional myoelectric control of wrist movements. And got the results that the kernel ridge regression outperformed the other methods, b
Data Loading...
