Adaptive neural & fuzzy controller for exoskeleton gait pattern control based on musculoskeletal modeling

In recent technological advancements, designing of exoskeleton have gained attraction from various scientists and researchers across the world. Exoskeletons are assistive devices which are used by the subjects with limb impairments. Various trajectory tracking control schemes have been developed in the past, however there are several challenges like human machine interaction, external disturbances and model uncertainties present that should be handled in the efficient designing of a controller. This paper deals with the development of 2-link lower limb extremity manipulator for rehabilitation purposes in bio-medical engineering. A novel control strategy is proposed to track the desired trajectory for assistive devices using the robust adaptive sliding mode controller coupled with Radial Basis Function (RBF) neural network and fuzzy techniques for function approximation. The simulations are carried out on a musculoskeletal model Gait2354 which can be scaled according to the physical dimensions of the experimental subject developed in OpenSim software. The controller designed is robust in terms of uncertainties and disturbances present in the system and reduces chattering problems. In addition, Minimum Learning Parameter is also incorporated in the system which reduces the number of parameters to only one, thereby reducing the computational cost and enhancing the real time performance of the system. For validation of the proposed model, Matlab Simulink environment is used. The different performance index parameters used in comparative analysis are Integral Absolute Error (IAE), Integral Square Error (ISE), Integral Time Absolute Error (ITAE) and Integral Time Square Error (ITSE). The simulation results show that SMC-RBF network has better performance and tracking rate as compared to other controllers implemented. The results obtained were also compared with the existing method to prove their effectiveness. The simulation results obtained ensures to provide effective and comfortable rehabilitation services for lower limb extremity in the future.