Perancangan Instrumen Sistem Navigasi Dengan Kalman Filter Dan Algoritma Smoothing Savitzky-Golay

Maliki, Maulana (2018) Perancangan Instrumen Sistem Navigasi Dengan Kalman Filter Dan Algoritma Smoothing Savitzky-Golay. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
07111340000018-Undergraduate_Theses.pdf - Published Version
Restricted to Repository staff only

Download (6MB) | Request a copy

Abstract

INS merupakan sistem navigasi berbasis inersia yang paling banyak digunakan, terutama dalam navigasi wahana tak berawak seperti UAV, USV, Rudal, dll. Namun data keluaran yang dibaca oleh Accelerometer dan gyroscope mengandung noise yang berakumulasi seiring waktu dikarenakan proses integral. Untuk mengurangi dan menghilangkan perngaruh noise pada data keluaran sensor navigasi, digunakan filter Kalman sebagai estimator noise. Integrasi menggunakan GPS juga digunakan guna menyediakan sinyal referensi untuk INS, sedangkan algoritma smoothing Savitzky-Golay digunakan untuk memperhalus keluaran dari Kalman filter. Hasil dari simulasi menunjukkan bahwa Kalman filter telah berhasil mengestimasi noise dari sinyal INS, meski masih ada mean error namun nilainya tergolong kecil, yakni [0.0323, 0.0323, 0.0329] m untuk error posisi, [0.0231, 0.0233, 0.0241] untuk error kecepatan dan [0.025, 0.0081, 0.0224] untuk error orientasi masing-masing pada sumbu [x,y,z]. ================= INS is an inertial-based navigation system which used in so many aplication such as unmanned vehicle like UAV, USV, missile, etc. But the noise in the output of inertial sensors accelerometer and gyroscope tend to grow whitout bound, this was caused by integration algorithm in mathematical process in INS. Kalman fitler was used as noise estimator to minimize the error. Other navigation system such GPS was also used as aiding to INS for providing a reference signal, so that we can estimate the error model, and lastly the Savitzky-Golay smoothing algorithm was used to smooth and minimize the error even further. From the simulation, the kalman filter succeed to reduce the error caused by noise signal, the error between kalman estimation and signal was [0.0323, 0.0323, 0.0329] m for position, [0.0231, 0.0233, 0.0241] m/s for velocity and [0.025, 0.0081, 0.0224] rad for attitude, all of it in [x,y,z] axis.

Item Type: Thesis (Undergraduate)
Additional Information: RSE 629.831 2 Mal p-1
Uncontrolled Keywords: INS, Navigasi, GPS, Kalman filter, noise, Savitzky-Golay Smoothing
Subjects: Q Science > QA Mathematics > QA402.3 Kalman filtering.
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing.
Divisions: Faculty of Electrical Technology > Electrical Engineering > (S1) Undergraduate Theses
Depositing User: Maulana Maliki
Date Deposited: 04 Dec 2018 05:06
Last Modified: 04 Dec 2018 05:06
URI: http://repository.its.ac.id/id/eprint/53218

Actions (login required)

View Item View Item