Kontrol Path Following Line pada Quadrotors UAV (Unmanned Aerial Vehicle) berbasis SMC-RBFNN dengan Gangguan Eksternal

Sari, Delima Palwa (2023) Kontrol Path Following Line pada Quadrotors UAV (Unmanned Aerial Vehicle) berbasis SMC-RBFNN dengan Gangguan Eksternal. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6009211014-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (6MB) | Request a copy

Abstract

UAV (Unmanned Aerial Vehicle) dapat dikendalikan dari jauh oleh pilot atau dapat dikendalikan secara autonomous. Jenis UAV quadrotor,saat ini banyak diteliti, karena memiliki beberapa keuntungan seperti kemampuan take off dan landing secara vertikal, dapat take off dan landing pada area yang tidak luas. Untuk meminimalkan terjadinya kesalahan pengoperasian oleh pilot ketika melakukan kontrol pada UAV, maka dikembangkan autonomous UAV. Pada penelitian ini, kinerja sliding mode control (SMC) dipadukan dengan radial basis function neural network (RBFNN) sebagai kontrol utama sistem dinamik quadrotor, yang diuji pada lintasan uji dengan gangguan eksternal melalui simulasi. Simulasi dilakukan dengan tiga lintasan uji dengan tanpa dan adanya gangguan eksternal. Gangguan eksternal diimplementasikan untuk koordinat x, y, z secara bersamaan. Hasil simulasi rata-rata nilai error pada berbagai lintasan uji untuk metode SMC-RBFNN dengan kondisi tanpa gangguan eksternal adalah 0,95 m dan metode SMC yaitu 2,08 m. Selanjutnya, pada kondisi dengan gangguan step, rata-rata nilai error metode SMC-RBBFNN adalah 1,87 m; untuk metode SMC adalah 4,05 m. Serta, pada kondisi dengan gangguan sinus, rata-rata nilai error metode SMC-RBBFNN adalah 1,16 m; untuk metode SMC adalah 4,02 m. Setelah dilakukan eksperimen hardware didapatkan hasil rata-rata nilai error pada berbagai lintasan uji untuk metode SMC-RBFNN dengan kondisi tanpa gangguan eksternal adalah 1,23 m dan metode SMC yaitu 1,80 m. Selanjutnya, pada kondisi dengan gangguan, rata-rata nilai error metode SMC-RBBFNN adalah 1,53 m; untuk metode SMC adalah 2,04 m. Sehingga dapat disimpulkan bahwa metode SMC-RBFNN merupakan salah satu strategi pengendalian yang tepat untuk misi penerbangan quadrotor UAV pada simulasi maupun hardware.
=====================================================================================================================================
UAV (Unmanned Aerial Vehicle) can be remotely controlled by a pilot or can be controlled autonomously. The quadrotor UAV type is currently being widely studied, because it has several advantages such as the ability to take off and land vertically, can take off and land in a small area. To minimize the occurrence of operating errors by pilots when controlling the UAV, an autonomous UAV was developed. In this study, the sliding mode control (SMC) performance was combined with the radial basis function neural network (RBFNN) as the main control for the quadrotor dynamic system, which was tested on a test track with external disturbances through simulation. The simulation is carried out with three test tracks with no and no external disturbances. External noise is implemented for x, y, z coordinates simultaneously. The simulation results mean the error value on various test paths for the SMC-RBFNN method with conditions without external interference is 0,95 m and for the SMC method is 2,08 m. Furthermore, in conditions with step disturbances, the average error value of the SMC-RBBFNN method is 1,87 m; for the SMC method is 4,05 m. Also, in conditions with sine faults, the average error value of the SMC-RBBFNN method is 1,16 m; for the SMC method is 4,02 m. After hardware experiments, the average error value on various test paths for the SMC-RBFNN method with conditions without external disturbance is 1,23 m and for the SMC method is 1,80 m. Furthermore, in conditions with disturbances, the average error value of the SMC-RBBFNN method is 1,53 m; for the SMC method is 2,04 m. So it can be concluded that the SMC-RBFNN method is one of the appropriate control strategies for quadrotor UAV flight missions in both simulation and hardware.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Quadrotor, UAV, Sliding Mode Controller (SMC), Radial Basis Function Neural Network (RBFNN).
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.8 Vehicles, Remotely piloted. Autonomous vehicles. T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL776 .N67 Quadrotor helicopters--Automatic control U Military Science > UG1242 Drone aircraft--Control systems. (unmanned vehicle)
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30101-(S2) Master Thesis
Depositing User:	Delima Palwa Sari
Date Deposited:	28 Aug 2023 01:37
Last Modified:	28 Aug 2023 01:37
URI:	http://repository.its.ac.id/id/eprint/101004

Actions (login required)

View Item