Fuzzy modelling of knee joint with genetic optimization

Issue title: Assistive and Rehabilitation Robotics

Article type: Research Article

Authors: Ibrahim, B.S.K.K. | Tokhi, M.O. | Huq, M.S. | Jailani, R. | Gharooni, S.C.

Affiliations: Department of Automatic Control and System Engineering, University of Sheffield, Sheffield, United Kingdom

Note: [] Corresponding author: Babul Salam Ksm Kader Ibrahim. E-mail: [email protected], [email protected]

Abstract: Modelling of joint properties of lower limbs in people with spinal cord injury is significantly challenging for researchers due to the complexity of the system. The objective of this study is to develop a knee joint model capable of relating electrical parameters to dynamic joint torque as well as knee angle for functional electrical stimulation application. The joint model consists of a segmental dynamic, time-invariant passive properties and uncertain time-variant active properties. The knee joint model structure comprising optimised equations of motion and fuzzy models to represent the passive viscoelasticity and active muscle properties is formulated. The model thus formulated is optimised using genetic optimization, and validated against experimental data. The developed model can be used for simulation of joint movements as well as for control development. The results show that the model developed gives an accurate dynamic characterisation of the knee joint.

Keywords: Knee joint model, viscoelasticity, functional electrical stimulation, fuzzy model, genetic algorithm

DOI: 10.3233/ABB-2011-0007

Journal: Applied Bionics and Biomechanics, vol. 8, no. 1, pp. 85-99, 2011