Perancangan Sistem Estimasi Posisi Traktor Berbasis Gps Dengan Algoritma Extended Kalman Filter

Azzuhaili, Muhammad Wahbah (2025) Perancangan Sistem Estimasi Posisi Traktor Berbasis Gps Dengan Algoritma Extended Kalman Filter. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5009211032-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Pertanian merupakan sektor penting dalam ekonomi Indonesia. Traktor otonom dapat menjadi solusi untuk meningkatkan produktivitas di sektor ini. Namun, sebelum traktor otonom dapat diimplementasikan secara efektif, diperlukan sistem estimasi posisi yang akurat. Estimasi posisi menjadi komponen krusial karena menentukan ketepatan lokasi dan arah gerak kendaraan selama beroperasi. Penelitian ini bertujuan untuk merancang dan mensimulasikan sistem estimasi posisi pada traktor berbasis sensor GPS u-blox NEO-6M dengan pendekatan Extended Kalman Filter (EKF). Pengujian dilakukan dengan membandingkan data GPS u-blox NEO-6M terhadap data referensi dari GPS RTK yang memiliki akurasi tinggi. Hasil awal menunjukkan adanya bias sistematis pada sensor GPS u-blox, yang kemudian dikoreksi untuk meningkatkan akurasi. Setelah dilakukan koreksi bias hasil RMSE dari GPS u-blox terhadap GPS RTK adalah 1,044 meter. Kemudian setelah menggunakan EKF, lintasan hasil estimasi menunjukkan kesesuaian yang lebih baik terhadap lintasan referensi dengan nilai RMSE terkecil sebesar 0,610 meter. Selain peningkatan pada estimasi posisi, algoritma EKF juga berhasil menstabilkan estimasi orientasi/heading, yang sebelumnya fluktuatif. Penelitian ini menunjukkan bahwa metode EKF efektif untuk meningkatkan akurasi dan kestabilan sistem navigasi traktor otonom berbasis GPS u-blox berbiaya rendah.
====================================================================================================================================
Agriculture is a crucial sector in the Indonesian economy. Autonomous tractors can be a solution to increase productivity in this sector. However, before autonomous tractors can be effectively implemented, an accurate position estimation system is required. Position estimation is a crucial component because it determines the accuracy of the vehicle's location and direction of movement during operation. This study aims to design and simulate a position estimation system on a tractor based on the u-blox NEO-6M GPS sensor using the Extended Kalman Filter (EKF) approach. Testing was conducted by comparing GPS u-blox data to reference data from a high-accuracy RTK GPS. Initial results indicated a systematic bias in the u-blox GPS sensor, which was then corrected to improve accuracy. After bias correction, the RMSE of the u-blox GPS versus the GPS RTK was 1.044 meters. Then, after using the EKF, the estimated trajectory showed better agreement with the reference trajectory with the smallest RMSE value of 0.610 meters. In addition to improving position estimation, the EKF algorithm also successfully stabilized the orientation/heading estimation, which was previously fluctuating. This study shows that the EKF method is effective in improving the accuracy and stability of a low-cost GPS-based autonomous tractor navigation system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	estimasi posisi, traktor otonom, GPS, Extended Kalman Filter (EKF), position estimation, autonomous tractor, GPS, Extended Kalman Filter (EKF)
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.8 Vehicles, Remotely piloted. Autonomous vehicles. T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL798.N3 Global Positioning System.
Divisions:	Faculty of Industrial Technology > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Wahbah Azzuhaili
Date Deposited:	06 Aug 2025 06:26
Last Modified:	06 Aug 2025 06:26
URI:	http://repository.its.ac.id/id/eprint/126631

Actions (login required)

View Item