Passive and active assistive writing devices in suppressing hand tremor

Abstract

Patients with hand tremor disease frequently experience difficulties in performing their daily tasks, especially in handwriting activities. In order to prevent the ingestion of drugs and intervention of surgeries, a non-invasive solution was presented to improve their writing capabilities. In this study, there were two novel inventions of the hand-held device named as TREMORX and Active Assistive Writing Device (AAWD) with the approaches of passive and active elements respectively. For validation, the patient with tremor was assisted in using a normal pen and TREMORX to perform a handwriting task at the sitting and standing postures. For AAWD, the active suppressing element was the servo motor to control the hand tremor act on the writing tool tip and an accelerometer will measure the necessary parameters values for feedback control signal. The classic Proportional (P) controller and Proportional-Integral- Derivative (PID) were presented. The P controller was tuned with a meta-heuristic method by adjusting the parameters into several values to examine the response and robustness of the controller in suppressing the tremor. The evaluation was based on decreasing the coherence magnitude on the frequency response analysis. To optimise the performances, two types of Evolutionary Algorithms (EA) were employed which were Genetic Algorithm (GA) and Particle Swarm Optimisation (PSO). The optimisation techniques were integrated into the PID controller system to generate the optimum performances in controlling the tremor. For the simulation study, the parametric model representing the actual system of the AAWD was presented. The main objectives of this analysis were to determine the optimum value of PID parameters based on EA optimisation techniques. The determined parameters for both optimisations were then injected into the experimental environment to test and evaluate the performance of the controllers. The findings of the study exhibited that the PID controller for both EA optimisation provided excellent performances in suppressing the tremor signal act on the AAWD in comparison to the classic pure P controller. Based on the fitness evaluation, the GA optimisation significantly enhanced the PID controller performance compared to PSO optimisation. The handwriting performance using both TRREMORX and AAWD was recorded and from a visual justification, it showed that the quality of legibility was improved as compared with using normal handwriting devices. These outcomes provided an important contribution towards achieving novel methods in suppressing hand tremor by means of the invention of the handheld writing devices incorporated with intelligent control techniques.