Adaptive notch filter for pathological tremor suppression using permanent magnet linear motor

Abstract

Tremor is a progressive neurological disease involving involuntary rhythmic muscle movements and oscillations of the human body parts by alternating contractions of muscles. Tremor is a common symptom that can severely limit the independence of affected individuals and their ability to perform activities of daily living. In this paper, a permanent magnet linear motor (PMLM) and a notch filter with an adaptive frequency estimator are employed to suppress pathological tremor. The effects of undesired intrinsic forces exhibited by the use of PMLM were minimized by development and parameter identification of a dynamic force model. The experimental evaluation of a real-time adaptive tremor identification and suppression controller was performed in a human joints emulator using recorded signals from participants with Parkinson’s disease and essential tremor. We employed wavelet transform to analyze the tremor signals and the performance of the controller in the time-frequency domain. Experimental results showed an average tremor suppression of 30.51 dB (97.0%) and 13.89 dB (79.8%) in the first and second tremor frequency components, respectively. The average resistance induced by the orthotic mechanism against the voluntary motion was 0.36 N, and a 7.74 dB reduction with an active force compensator was observed. Compared to a servo-pneumatic actuator, a PMLM offers higher energy efficiency at a comparable suppression performance.