SMC-ANFIS untuk Kontrol Pelacakan Lintasan dan Penghindaran Rintangan pada Autonomous UAV Quadcopter

Indayu, Nor (2023) SMC-ANFIS untuk Kontrol Pelacakan Lintasan dan Penghindaran Rintangan pada Autonomous UAV Quadcopter. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6009211007-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Penelitian ini bertujuan untuk meningkatkan kinerja pelacakan lintasan dan penghindaran rintangan pada UAV quadcopter menggunakan pengendali adaptive neuro-fuzzy inference system (ANFIS) berdasarkan sliding mode control (SMC). Pengendali SMC-ANFIS menggabungkan aturan-aturan fuzzy dengan kontrol SMC untuk mengoptimalkan output kontrol dan memungkinkan adaptasi terhadap perubahan lingkungan. Simulasi dinamika quadcopter dilakukan dengan menguji pengendali SMC-ANFIS pada variasi kondisi, termasuk pelacakan lintasan dan penghindaran rintangan. Hasil simulasi menunjukkan bahwa pengendali SMC-ANFIS mengungguli pengendali SMC dalam hal pelacakan lintasan, dengan nilai maksimum RMSE posisi adalah 0,053 meter. Selain itu, pengendali SMC-ANFIS juga mampu meningkatkan kemampuan quadcopter dalam menghindari rintangan tanpa adanya tabrakan. Simulasi penghindaran rintangan dengan variasi jenis dan jarak antar rintangan menunjukkan bahwa pengendali SMC-ANFIS menghasilkan penghindaran yang aman dengan tingkat chattering yang lebih rendah dibandingkan dengan pengendali SMC. Namun, pengendali SMC-ANFIS memerlukan waktu eksekusi yang lebih lama dibandingkan dengan pengendali SMC, mengakibatkan respon yang sedikit lebih lambat dalam situasi yang membutuhkan tindakan cepat. Rata- rata waktu eksekusi yang dibutuhkan oleh SMC 28.59% lebih cepat dari SMC-ANFIS.
=================================================================================================================================
This research aims to enhance the performance of trajectory tracking and obstacle avoidance in a quadcopter UAV using an adaptive neuro-fuzzy inference system (ANFIS) controller based on sliding mode control (SMC). The SMC-ANFIS controller combines fuzzy rules with SMC to optimize control outputs and enable adaptation to environmental changes. Quadcopter dynamics simulations were conducted by testing the SMC-ANFIS controller under various conditions, including trajectory tracking and obstacle avoidance. Simulation results demonstrate that the SMC-ANFIS controller outperforms the SMC controller in terms of trajectory tracking, with a maximum position root mean square error (RMSE) of 0.053 meters. Moreover, the SMC-ANFIS controller also enhances the quadcopter's obstacle avoidance capabilities, preventing collisions. Simulations of obstacle avoidance with varying obstacle types and distances show that the SMC-ANFIS controller achieves safe avoidance with lower chatter levels compared to the SMC controller. However, the SMC-ANFIS controller requires a longer execution time than the SMC controller, resulting in slightly slower respones in situations that demand quick actions. On average, the execution time needed by the SMC controller is 28.59% faster than the SMC-ANFIS controller.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	SMC-ANFIS, Quadcopter , Pelacakan Lintasan, Penghindaran Rintangan, Trajectory Tracking, Obstacle Avoidance
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL776 .N67 Quadrotor helicopters--Automatic control U Military Science > UG1242 Drone aircraft--Control systems. (unmanned vehicle) U Military Science > U Military Science (General) > UG Military Engineering > UG1242.D7 Unmanned aerial vehicles. Drone aircraft
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30101-(S2) Master Thesis
Depositing User:	Nor Indayu
Date Deposited:	05 Sep 2023 05:09
Last Modified:	05 Sep 2023 05:09
URI:	http://repository.its.ac.id/id/eprint/101122

Actions (login required)

View Item