The number of people with spinal cord injury (SCI) is increasing every year and walking has been found to be the most exciting and important prospect to these patients to improve their quality of life. Many individuals with incomplete SCI have the potential to walk and everyone of them wants to try. Unfortunately up to now, there is less than one third of patients could walk again after SCI. Residual function, the orthotic support, energy expenditure, patient motivation and control technique are some of the factors that influence the walking outcome of spinal cord injured people. In this thesis, a series of studies are carried out to investigate the possibility of enhancing the performance of the functional electrical stimulation (PES) assisted paraplegic walking with wheel walker through the development and implementation of intelligent control technique and spring brake orthosis (SBO) with full utilization of the voluntary upper body effort. The main aim of this thesis is to enable individuals with complete paraplegia to walk again with maximum performance and the simplest approach as possible. Firstly, before simulation of the system can be made, it is important to select the right model to represent the actual plant. In this thesis, the development of a humanoid and wheel walker models are carried out using MSC.visualNastran4D (vN4D) software and this is integrated with Matlab Simulink® for simulation. The newly developed quadriceps and hamstrings muscle models from the series of experiments are used to represent subject muscles after comparison and validation with other two well-known muscle models are performed. Several experiments are conducted to investigate the effect of stimulation frequency and pulse-width in intermittent stimulation with isometric measurement from paraplegic subjects. The results from this work can serve as a guidance to determine the optimum stimulation parameters such as frequency and pulse-width to reduce muscle fatigue during PES application. The ability test is introduced to determine the maximum leg force that can be applied to the specific paraplegic subject during FES functional task with minimum chance of spasm and leg injury. Investigations are carried out on the control techniques implemented for FES walking with wheel walker. PID control and fuzzy logic control (FLC) are used to regulate the electrical stimulation required by the quadriceps and hamstrings muscles in order to perform the FES walking manoeuvre according to predefined walking trajectory. The body weight transfer is introduced to increase the efficiency of FES walking performance. The effectiveness of body weight transfer and control strategy to enhance the performance of FES walking and reduce stimulation pulses required is examined. Investigations are carried out on the effectiveness of spring brake orthosis (SBO) for FES assisted paraplegic walking with wheel walker. A new concept in hybrid orthotics provides solutions to the problems that affect current 'hybrid orthosis, including knee and hip flexion without relying on the withdrawal reflex or a powered actuator and foot-ground clearance without extra upper body effort. The use of SBO can also eliminate electrical stimulation pulses required by the hamstrings muscle for the same FES walking system. Further improvement of the FES walking system is achieved by introducing finite state control (FSC) to control the switching time between springs, brakes and electrical stimulation during FES assisted walking with wheel walker with the combInation of FLC to regulate the electrical stimulation required for the knee extension. The results show that FSC can be used to accurately control the switching time and improve the system robustness and stability.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:538098 |
| Date | January 2011 |
| Creators | Jailani, Rozita |
| Publisher | University of Sheffield |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://etheses.whiterose.ac.uk/14989/ |
Page generated in 0.0042 seconds