Square Root Ensemble Kalman Filter for Forefinger Motion Estimation as Post-Stroke Patients’ Medical Rehabilitation

T. Herlambang, F. A. Susanto, H. Nurhadi, K. Oktafianto, H. Arof, R. S. Marjianto

Research output: Contribution to journalArticlepeer-review

Abstract

Hemiparesis is a medical term to describe a condition of weakness on one side of the body or the inability to move a limb on one side. The term comes from the word’hemi’ meaning half or one side and “paresis” meaning weakness. Hemiparesis patients are still able to move the affected side of the body and are not completely paralyzed. It is just that the side of the body experiencing the disorder is so weak and powerless, the movements that arise are also very little. Current robotics technology has developed rapidly along with advances in science and technology to assist medical rehabilitation, one of which is a post-stroke rehabilitation, especially the rehabilitation of finger movement. One technology useful to develop is the finger motion estimation of the Finger Prosthetic Robotic Arm for patients with upper extremity paresis. Finger motion estimation is one of the important aspects in the development of such technology because it is designed to determine the accuracy and effectiveness of the robot in providing motion assistance to hands affected by paresis. The study in this paper used the Square Root Kalman Filter method to estimate the motion of the index finger robot by generating 250 ensembles, 500 ensembles and 750 ensembles. The simulation results with 750 ensembles have the best accuracy of around 97-98%.

Original languageEnglish
Pages (from-to)392-399
Number of pages8
JournalNonlinear Dynamics and Systems Theory
Volume24
Issue number4
Publication statusPublished - 2024

Keywords

  • EnKF
  • SR-EnKF
  • forefinger motion estimation
  • hemiparesis, finger prosthetic robotic arm

Fingerprint

Dive into the research topics of 'Square Root Ensemble Kalman Filter for Forefinger Motion Estimation as Post-Stroke Patients’ Medical Rehabilitation'. Together they form a unique fingerprint.

Cite this