Modelling and control of paraplegic's knee joint (FES-swinging)

• Main Author: Kader Ibrahim, Babul Salam
• Published: University of Sheffield 2011
• Subjects: 616.842
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.556679

The use of electrical signals to restore the function of paralyzed muscles is called functional electrical stimulation (PES). PES is a promising method to restore mobility to individuals paralyzed due to spinal cord injury (SCI). This thesis is concerned with the development of an accurate paraplegic knee joint model and control of electrically stimulated muscle. The modelling of musculoskeletal of paraplegic's lower limb is significantly challenging due to the complexity of the system. The first aim 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 PES application. The knee joint is divided into 3 parts; active muscle properties, passive knee joint properties and lower limb dynamics Hence the 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 results show that the model developed gives an accurate dynamic characterisation of the knee joint. The second aim of this study is to develop PES-induced swinging motion control. A crucial issue of PES is the control of motor function by artificial activation of paralyzed muscles. Major problems that limit the success of current PES control systems are nonlinearity of the musculoskeletal system and rapid change of muscle properties due to fatigue. Fuzzy logic control (FLC) with its ability to handle a complex nonlinear system without mathematical model is used. Two FLC strategies; trajectory based control and cycleto- cycle control are developed. In the trajectory based control, the controller with less energy consumption is developed to reduce muscle fatigue. The ability of this controller to minimize the fatigue is proved in the experimental work. The discretetime cycle-to-cycle control strategy is developed without predefined trajectory. This strategy is applicable for controlling PES-induced movement with the ability to reach full knee extension angle and to maintain a steady swinging of the lower limb as desired in the presence of muscle fatigue and spasticity. The performances of the controllers are assessed through simulation study and validated through experimental work.