Desain Sistem Hexacopter Berbasis Model Predictive Control untuk Pembersihan Kaca Gedung Tinggi

Suntani, Pijar Patria (2026) Desain Sistem Hexacopter Berbasis Model Predictive Control untuk Pembersihan Kaca Gedung Tinggi. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221122-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Pembersihan fasad kaca pada gedung bertingkat merupakan pekerjaan berisiko tinggi dan tidak efisien jika dilakukan secara manual. Penggunaan UAV (Unmanned Aerial Vehicle) jenis hexacopter memberikan solusi otomasi yang menjanjikan. Namun, operasi pembersihan ini menghadapi tantangan mekanis berupa gaya dorong balik (recoil force) dan torsi pitch yang dihasilkan oleh pancaran air bertekanan tinggi. Gangguan dinamis ini dapat mengancam kestabilan penerbangan, sementara metode kontrol konvensional sering kali kurang memadai dalam menangani dinamika multivariabel (MIMO) secara bersamaan. Mengatasi batasan tersebut, penelitian ini mengusulkan perancangan sistem kontrol yang mengintegrasikan Model Predictive Control (MPC) dan Finite State Machine (FSM). MPC bertugas mengkompensasi gangguan recoil dan menjaga kestabilan orientasi hexacopter, sementara FSM bertindak sebagai pembangkit lintasan otonom untuk menyinkronkan manuver sapuan dengan aktivasi pompa. Melalui simulasi MATLAB, kinerja sistem dievaluasi dengan membandingkan arsitektur Coupled dan Decoupled. Hasil pengujian membuktikan bahwa arsitektur Decoupled memberikan kinerja paling optimal. Sistem terbukti mampu melacak lintasan secara presisi dengan Root Mean Square Error (RMSE) sumbu vertikal sebesar 1,6872 meter tanpa memicu fenomena overshoot, serta berhasil meredam fluktuasi sudut pitch secara instan. Konfigurasi ini juga mencatat efisiensi operasional terbaik dengan menyelesaikan misi bermanuver dalam durasi tercepat yakni 140,60 detik, secara efektif menyisakan kapasitas baterai sebesar 98,88%, dan mempertahankan sisa cairan pembersih sebanyak 2,44 liter.
=================================================================================================================================
Cleaning the glass facades of high-rise buildings is a high-risk and inefficient task when performed manually. The use of an Unmanned Aerial Vehicle (UAV) of the hexacopter type provides a promising automation solution. However, this cleaning operation faces mechanical challenges in the form of recoil force and pitch torque generated by high-pressure water jets. These dynamic disturbances can threaten flight stability, while conventional control methods are often inadequate in handling multivariable dynamics (MIMO) simultaneously. To overcome these limitations, this study proposes the design of a control system integrating Model Predictive Control (MPC) and a Finite State Machine (FSM). The MPC is tasked with compensating for the recoil disturbance and maintaining the orientation stability of the hexacopter, while the FSM acts as an autonomous trajectory generator to synchronize the sweeping maneuvers with pump activation. Through MATLAB simulations, the system's performance was evaluated by comparing Coupled and Decoupled architectures. The test results prove that the Decoupled architecture provides the most optimal performance. The system is proven capable of tracking the trajectory precisely with a Root Mean Square Error (RMSE) on the vertical axis of 1,6872 meters without triggering overshoot phenomena, and successfully dampens pitch angle fluctuations instantly. This configuration also records the best operational efficiency by completing the maneuvering mission in the fastest duration of 140,60 seconds, effectively leaving a battery capacity of 98,88%, and maintaining the remaining cleaning fluid at 2.44 liters.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Finite State Machine (FSM), Gaya Tolak Mundur, Hexacopter, Model Predictive Control (MPC), Pembersihan Fasad Kaca pada Gedung, Facade Cleaning, Finite State Machine (FSM), Hexacopter, Model Predictive Control (MPC), Recoil Force
Subjects:	T Technology > T Technology (General) > T11 Technical writing. Scientific Writing T Technology > T Technology (General) > T57.62 Simulation
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Pijar Patria Suntani
Date Deposited:	24 Jul 2026 07:32
Last Modified:	24 Jul 2026 07:32
URI:	http://repository.its.ac.id/id/eprint/137088

Actions (login required)

View Item