Perancangan Dan Evaluasi Navigasi Drone Otonom Berbasis RFID Pada Perkebunan Kelapa Sawit Menggunakan ROS

Dwihar, Savitri Rizquna Azzahra (2026) Perancangan Dan Evaluasi Navigasi Drone Otonom Berbasis RFID Pada Perkebunan Kelapa Sawit Menggunakan ROS. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5009221128-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Perkebunan kelapa sawit memiliki area yang luas dan jumlah tanaman yang banyak, sehingga diperlukan sistem pemantauan yang efisien dan mampu melakukan identifikasi tanaman secara individual. Pemanfaatan drone otonom menjadi salah satu solusi potensial, namun dibutuhkan sistem navigasi yang mampu bergerak secara otonom, mengidentifikasi target, dan menghindari rintangan di lingkungan perkebunan. Pada Tugas Akhir ini dikembangkan sistem navigasi pada drone otonom berbasis RFID dengan Robot Operating System (ROS) sebagai framework pengembangan sistem. ROS berperan sebagai penghubung antar subsistem, termasuk navigasi, deteksi RFID, perencanaan jalur, dan kontrol penerbangan. Gazebo digunakan sebagai simulator fisik, sementara ArduPilot digunakan sebagai autopilot dengan kontrol PID yang telah tersedia untuk menjaga stabilitas quadcopter. Metode yang digunakan mengintegrasikan deteksi RFID dengan metode lokalisasi Synthetic Aperture Radar (SAR) sebagai sumber informasi posisi target dengan algoritma Rapidly-exploring Random Tree (RRT) sebagai metode obstacle avoidance. Pada tahap simulasi, quadcopter diarahkan dari homebase menuju RFID tag 1 di posisi (10,0), dilanjutkan ke RFID tag 2 di posisi (10,10), menghindari tiga rintangan di sepanjang lintasan dan kembali ke homebase dengan ketinggian konstan 5 meter. Pengulangan dilakukan sebanyak lima kali untuk mengevaluasi performa sistem. Hasil simulasi menunjukkan bahwa sistem RFID memiliki nilai RMSE sebesar 0,116 meter. Selain itu, lintasan pergerakan quadcopter menghasilkan nilai RMSE sebesar 0,070 meter dan maksimum error lintasan sebesar 0,311 meter. Performa kontrol PID menunjukkan respon sistem yang stabil dengan rise time sebesar 3 detik, settling time sebesar 7,3 detik, dan overshoot sebesar 0,8% pada pengujian takeoff.
================================================================================================================================
Oil palm plantations cover large areas and consist of a high number of plants, thereby requiring an efficient monitoring system capable of performing individual plant identification. The use of autonomous drones represents a potential solution; however, it requires a navigation system that can operate autonomously, identify targets, and avoid obstacles within the plantation environment. In this final project, an adaptive navigation system for an RFID-based autonomous drone is developed using the Robot Operating System (ROS) as the system development framework. ROS serves as an interface connecting various subsystems, including navigation, RFID sensing, path planning, and flight control. Gazebo is utilized as a physical simulation environment, while ArduPilot is employed as the autopilot with an integrated PID control scheme to maintain quadcopter stability. The proposed method integrates RFID detection with Synthetic Aperture Radar (SAR) based localization as the source of target position information combined with the Rapidly-exploring Random Tree (RRT) algorithm for obstacle avoidance. During the simulation stage, the quadcopter is guided from the home base to RFID tag 1 located at (10,0), followed by RFID tag 2 at (10,10), while avoiding three obstacles along the trajectory and returning to the home base at a constant altitude of 5 meters. The simulation is repeated five times to evaluate the system’s performance. The simulation results indicate that the RFID system achieves RMSE of 0,116 meter. Furthermore, the quadcopter trajectory yields an RMSE of 0.070 meters and a maximum path error of 0.311 meters. The PID control performance demonstrates stable system behavior, with a rise time of 3 seconds, a settling time of 7.3 seconds, and an overshoot of 0.8% during the takeoff test.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Drone otonom, Navigasi, RFID, ROS
Subjects:	T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL776 .N67 Quadrotor helicopters--Automatic control
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Savitri Rizquna Azzahra Dwihar
Date Deposited:	03 Feb 2026 00:52
Last Modified:	03 Feb 2026 00:52
URI:	http://repository.its.ac.id/id/eprint/131758

Actions (login required)

View Item