Tremor Suppression Using Internal Model Principle

Traditional treatments, such as medications and deep brain stimulation, are commonly used to manage tremors. However, these methods often involve risks and side effects, creating a need for non-invasive alternatives. According to control theory, the Internal Model Principle (IMP) can perfectly track or reject known periodic signals. Since tremors consist of three harmonic (sinusoidal) components, IMP is a suitable approach for cancelling them without affecting non-periodic components like voluntary motion. In this work, we propose a control method based on the IMP to suppress tremors while preserving voluntary motion. The method uses three IMP controllers and a proportional (K<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">P</inf>) controller to target the harmonic components of the tremor. The controller parameters are calculated offline using the open-loop transfer function and the geometric average of each tremor bandwidth. Simulation results and experimental testing with real patient tremor data demonstrate that the proposed method effectively cancels tremor signals while preserving voluntary motion. Across 10 patient datasets, it achieved a median tremor suppression of 85.3 % and voluntary motion preservation of 85 %. The method has a low computational cost and a simple structure, making it suitable for future use in wearable systems such as tremor suppression gloves.