Rate-Dependent Gait Dynamic Stability Analysis for Motor Control Estimation

Abstract

This work presents the gait dynamic stability study for different walking speeds adopted by user intentions. Experimental data were collected from four healthy subjects while walking on a force platform at slow, normal and fast speed. The rate-dependent variations in the center of pressure (COP) and ground reaction forces (GRF) were modelled as motor output and input responses. Finite difference and non-linear regression algorithms were implemented to model gait transitions. Dynamic stability estimation for level ground walking was performed by analysis in time and frequency domains. Study of the COP velocity in loading phase showed that, the overdamped motor output response acts as a compensator for instabilities and oscillations in unloading phase and initial contact. Normal walking was predicted, from gait analysis in frequency domain, as the most stable gait for healthy subjects.

Towards Behavioral Based Sensorimotor Controller Design for Wearable Soft Exoskeletal Applications

“…The balancing approach adopted by sensorimotor may be estimated by behavioral I/O's response exhibited by the human exoskeletal system. The frequency domain analysis and estimation of motor control presented in this work extend previous studies on stability analysis in time domain [1][2][3][4][5][6].…”

“…In this study, the neuromuscular dynamic stability thresholds are modelled in terms of rate dependent variations in the whole body COM and the COP signals. A disturbance in the COM acceleration is modelled as motor input, while the COP velocity is modelled as motor output, which corresponds to the muscular response via spine [5,6]. The correlation between two signals was used to determine an exoskeletal model (plant) and the neuromotor behavioral estimation (controller) as shown in Fig.…”

“…Imran Mahmood, Uriel Martinez-Hernandez and Abbas A. Dehgani-Sanij are with the Institute of Design, Robotics and Optimisation (iDRO), and the School of Mechanical Engineering at the University of Leeds, Leeds, is proposed using the affine parameterization method. This modelling approach has been successfully implemented for activities like level ground walk and ramp walk [5,6].…”