Neuro-sliding mode control with modular models for control of knee-joint angle using quadriceps electrical...

In this paper, we propose a control methodology which is based on synergistic combination of a single-neuron controller with sliding mode control (SMC) for control of knee-joint position in paraplegic subjects with quadriceps stimulation. The control law will be switched from the sliding mode control to neural control, when the state trajectory of system enters in some boundary layer around the sliding surface. The main drawback of the standard sliding modes is mostly related to the so-called chattering caused by the high-frequency control switching. The value of switching gain depends on the bounds of system uncertainties. The system with large uncertainties needs to use a higher switching gain. This will, however, result in the high-frequency control switching and chattering across the sliding surface. To avoid such a condition, it is necessary to decrease the system uncertainty. To decrease the uncertainty, an accurate model of the system is required. For this purpose, we present a modular approach to modeling the knee-joint dynamics. Extensive experiments on healthy and paraplegic subjects are provided to demonstrate the robustness, stability and tracking accuracy of the neuro-SMC. The experimental results show that the neuro-SMC provides excellent tracking control for different reference trajectories and could generate control signals to compensate the muscle fatigue.