Garuda - Garba Rujukan Digital

Khoirudin Fathoni, Khoirudin

Unknown Affiliation

Author-ID : 609958

Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Energy Engineering

Published : 17 Documents Claim Missing Document

Claim Missing Document

Articles

Title

1

Pengenal Gerakan dengan Joystick Akselerometer Menggunakan Filter Kalman Khoirudin Fathoni; Dhidik Prastiyanto
Jurnal Rekayasa Elektrika Vol 13, No 3 (2017)
Publisher : Universitas Syiah Kuala

Human Machine Interaction keeps growing and developing, one of development is through gesture recognition that detects acceleration in a movement. This technology has been applied in joystick Wiimote and Wii-nunchuk by Nintendo that is widely used all over the world. Two main challenges in using accelerometer are to eliminate the noise of the sensor and to cancel the detected gravity acceleration when the joystick is tilted. The noise and gravity acceleration may influence the data reading and create error accumulation, respectively. This work proposes an implementation of Kalman Filter and also a simple technique to eliminate the influence of the gravity acceleration as a solution to solve above problems in using accelerometer of Wii-Nunchuk joystick in Board Arduino Mega 2560. The experimental results in motionless position show that the filter can reduce the gravity acceleration. We have to set the initial value of q and R parameters in the estimation of position, speed, and acceleration using Kalman filter. Once R is decided, the change of q will determine Kk gain, and it will locate the poles of the observer that influence the stability and the estimation result. With R=0.00005 and q=1, the poles of Kalman filter are located in the unit circle so that the estimation is stable and appropriate with the data from the sensor and even cancel the noise.

Pengenal Gerakan dengan Joystick Akselerometer Menggunakan Filter Kalman Khoirudin Fathoni; Dhidik Prastiyanto
Jurnal Rekayasa Elektrika Vol 13, No 3 (2017)
Publisher : Universitas Syiah Kuala

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.17529/jre.v13i3.8570

Human Machine Interaction keeps growing and developing, one of development is through gesture recognition that detects acceleration in a movement. This technology has been applied in joystick Wiimote and Wii-nunchuk by Nintendo that is widely used all over the world. Two main challenges in using accelerometer are to eliminate the noise of the sensor and to cancel the detected gravity acceleration when the joystick is tilted. The noise and gravity acceleration may influence the data reading and create error accumulation, respectively. This work proposes an implementation of Kalman Filter and also a simple technique to eliminate the influence of the gravity acceleration as a solution to solve above problems in using accelerometer of Wii-Nunchuk joystick in Board Arduino Mega 2560. The experimental results in motionless position show that the filter can reduce the gravity acceleration. We have to set the initial value of q and R parameters in the estimation of position, speed, and acceleration using Kalman filter. Once R is decided, the change of q will determine Kk gain, and it will locate the poles of the observer that influence the stability and the estimation result. With R=0.00005 and q=1, the poles of Kalman filter are located in the unit circle so that the estimation is stable and appropriate with the data from the sensor and even cancel the noise.

Title Search

Found 2 Documents Search Journal : Jurnal Rekayasa elektrika

Abstract

Abstract

Title

Found 2 Documents
Search
Journal : Jurnal Rekayasa elektrika