Pengembangan Peralatan Programmable Fes System Dengan Wearable Sensor

Setiawan, Rachmad (2018) Pengembangan Peralatan Programmable Fes System Dengan Wearable Sensor. Doctoral thesis, Institut Teknologi Sepuluh Nopember.

Text 07111060010004-dissertation.pdf - Published Version Restricted to Repository staff only Download (4MB)

Abstract

Beberapa penyakit, seperti stroke dan spinal cord injury, dapat mengakibatkan kelumpuhan serta hilangnya fungsi dasar anggota gerak pada individu yang mengalaminya. Fungsi tersebut dapat dikembalikan melalui rehabilitasi medik menggunakan metode Functional Electrical Stimulation (FES). Agar rehabilitasi dapat berjalan efektif serta dapat meningkatkan kualitas hidup dan kemandirian individu, maka sistem FES harus dibuat dengan sistem kontrol closed-loop. Dibutuhkan dua parameter gait yang digunakan sebagai feedback, yaitu fase gait dan sudut persendian tungkai bawah. Penerapan sistem ini akan menimbulkan masalah yaitu pergerakan pasien yang terbatas dan faktor psikologis pasien yang terganggu akibat pemasangan sensor dan pengkabelan yang rumit. Oleh karena itu dibutuhkan sensor yang wearable. Pada penelitian ini, untuk mengukur sudut sendi hip, knee dan ankle digunakan sensor inersia yang terdiri dari akselerometer dan giroskop. Sedangkan untuk mendeteksi fase gait digunakan force sensing resistor (FSR) dengan metode theresholding. Karena sensor giroskop mengalami dc error drift offset sedang sensor akselerometer mengalami gangguan karena gerakan translasi, maka diperlukan Filter Kalman untuk estimasi sudut tilt tiap segmen. Semua proses dilakukan secara real time oleh sebuah mikrokontroler 16-bit, dsPIC4013. Data hasil pengukuran dikirim ke PC secara nir-kabel lewat Bluetooth. Stimulator FES yang bisa diatur durasi dan amplitudonya akan lebih bisa diterima oleh pasien. Dari hasil pengujian akselerometer, digunakan pre-processing dengan filter digital Butterworth LPF order 2, dengan frekuensi cut-off 4Hz. Koefisien A = 0,013359; B=0,026718; C=0,013359; D=1,647462; dan E=-0,700899. Sedangkan dari pengujian sistem dengan model kaki statis, didapat rata-rata RMSE sudut sendi hip = 0,32o; sendi knee = 0,56o; sendi ankle = 0,83o. Dengan nilai RMSE ini, joint angles hasil Filter Kalman mewakili nilai aktual. Hasil perbandingan data pengukuran pergerakan lower limb joint angles dengan data penelitian yang sudah ada, menunjukkan pola sinyal yang sama, dengan magnitudo masih berada dalam nilai standard deviasi data pembanding. Frekuensi FES = 337 Hz dengan duty cycle mulai dari 1% sampai 14%, dengan nilai amplitudo linier dari 22.6V sampai 133.4V. Desain FES dan sistem pengukuran joint angles ini diharapkan bisa digunakan untuk penerapan beberapa metode kontrol untuk rehabilitasi lower limb karena durasi dan amplitudonya yang bisa diatur secara programmable. Rencana selanjutnya ditujukan untuk menerapkan stimulator yang dikembangkan dan diuji dalam penelitian ini ke dalam metode siklus siklus kontrol untuk meningkatkan kinerja kontrolnya.
============================================================
Some diseases, such as stroke and spinal cord injury, can result in paralysis and loss of basic limb function of an individual who experiences it. The function can be restored through medical rehabilitation using Functional Electrical Stimulation (FES) method. In order for rehabilitation to work effectively and to improve the quality of life and individual independence, the FES system must be designed in a closed-loop control system. Threre are two gait parameters used as feedback, i.e. gait phase and lower limb joint angle. Nevertheless, the implementation of this system will cause problems when patient's motion is limited, and the psychological factors of the patient are impaired due to the installation of sensors and the complicated wiring. Therefore, a wearable sensor is required. In this study, to measure the joint angle of hip, knee and ankle, inertial sensors consisting of accelerometers and gyroscopes are used. While to detect gait phase, force sensing resistors (FSRs) are used with a theresholding method. Since the gyroscope sensor experiences dc error drift offset, while the accelerometer sensor is interrupted due to translational motion, Kalman Filter is required to estimate the tilt angle of each segment. In the system implementation, all processes are performed in real time by a 16-bit microcontroller, dsPIC4013. Measurement data are sent to PC wirelessly via Bluetooth. FES Stimulator, that the duration and amplitude can be adjusted, will be more acceptable and pleasant to the patient. From accelerometer test result, pre-processing with digital filter of second-order Butterworth LPF is used, with cut-off frequency of 4Hz. The coefficients are A = 0,013359; B = 0.026718; C = 0.013359; D = 1.647462; and E = -0,700899. While from the system test with static foot model, the average RMSE joint angle are hip = 0.32o; knee joint = 0.56o; ankle joint = 0.83o. With a relatively small RMSE value, the joint angles of the Kalman Filter represent the actual values. The result of comparison of measurement data of lower limb joint angle movement with existing research data shows the same signal pattern, with magnitude still in the standard deviation value of comparison data. Frequency FES = 337 Hz with duty cycle ranging from 1% to 14%, with linear amplitude values from 22.6 V to 133.4 V. The designs of FES and joint angles measurement system are expected to be used for the application of several control methods for rehabilitation of lower limb since the duration and amplitudes can be easily adjusted based on designed control method.

Item Type:	Thesis (Doctoral)
Additional Information:	none
Uncontrolled Keywords:	lower limb joint angles, gait, sensor inersia, Kalman filter, Functional Electrical Stimulation, lower limb joint angles, gait, inertia sensor, Kalman filter, Functional Electrical Stimulation
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Industrial Technology > Electrical Engineering > 20001-(S3) PhD Thesis
Depositing User:	ansi aflacha
Date Deposited:	28 Nov 2018 02:24
Last Modified:	28 Nov 2018 02:24
URI:	http://repository.its.ac.id/id/eprint/60030

Actions (login required)

View Item