Robust Designer for Rehabilitation Robots Against Noise and Parametric Uncertainty

Robust Designer for Rehabilitation Robots Against Noise and Parametric Uncertainty

Abbas Nemati1 Shabnam Pourmasoud2

1) Electrical Engineering Department, Islamic Azad University, Mianeh Branch, Mianeh, Iran.
2) Master of Science in Electrical Engineering, Integrated Electronic Circuits, Islamic Azad University, Mianeh Branch, Mianeh, Iran.

IOI Code : AAHD-ABGAHD
Publication : National Conference on Information Technology, Nanotechnology, Artificial Intelligence and Technological Futures Studies(itnaf.ir)
Abstract :
    Rehabilitation robots have gained significant attention in recent years as important tools for the motor recovery of patients. The accurate performance of these robots depends on the design of advanced control systems capable of ensuring precise and safe motion in the presence of measurement noise, external disturbances, and parametric uncertainties. However, many conventional control methods face limitations when dealing with the nonlinear dynamics of robotic systems and non‑ideal environmental conditions. Therefore, the development of robust and intelligent control strategies that can guarantee tracking accuracy and system stability under real-world conditions is of great importance. The main objective of this study is to design and propose a novel control architecture to improve the performance of rehabilitation robots in the presence of noise and parametric uncertainties. In this research, a multilayer robust neural–predictive controller is proposed for a rehabilitation robot. In the proposed structure, measurement noise is first estimated using a predictive model based on an autoregressive process, and its effect on the control input signals is reduced. Then, a neural network is employed as a nonlinear approximator to estimate dynamic uncertainties and unknown parts of the system model. Finally, a nonlinear robustifying component based on a hyperbolic tangent function is incorporated into the control law to maintain system stability while preventing severe control oscillations. The stability of the control system is analyzed using Lyapunov stability theory. To evaluate the performance of the proposed method, several simulations were conducted in the MATLAB/Simulink environment, and the results were compared with benchmark controllers including PID, sliding mode control, and adaptive fuzzy control. The simulation results demonstrate that the proposed approach significantly improves trajectory tracking accuracy while reducing control signal oscillations. Moreover, the controller shows strong robustness against measurement noise and parametric variations, maintaining system stability under different operating conditions.

Keywords : rehabilitation robot robust control neural network noise prediction nonlinear control