Perancangan Kendali Haluan Untuk Menghindari Halangan Pada Kapal Menggunakan Model Predictive Control Berbasis Metode Kolokasi Chebyshev

Zulfikar, Zaky (2026) Perancangan Kendali Haluan Untuk Menghindari Halangan Pada Kapal Menggunakan Model Predictive Control Berbasis Metode Kolokasi Chebyshev. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002221036-Undergraduate-Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Penelitian ini bertujuan merancang kendali haluan kapal perang Extended-Corvette kelas SIGMA untuk menghindari halangan dinamis berupa kapal lain dengan mematuhi regulasi internasional COLREGS menggunakan Model Predictive Control (MPC) berbasis metode kolokasi Chebyshev. Dengan memanfaatkan model matematika 2 degree of freedom (sway dan yaw) serta pendekatan diskritisasi pseudospektral, algoritma ini dirancang untuk mengatasi tingginya beban komputasi pada MPC konvensional melalui transkripsi masalah kontrol optimal menjadi pemrograman kuadratik (Quadratic Programming) pada titik-titik Chebyshev-Gauss-Lobatto. Uji coba algoritma MPC Chebyshev dilakukan melalui simulasi menggunakan MATLAB yang mencakup 4 skenario uji navigasi, yaitu simulasi pada lintasan garis lurus tanpa halangan, situasi overtaking, head-on, dan crossing. Hasil pengujian menunjukkan bahwa algoritma perancangan jalur yang diusulkan bekerja secara presisi, di mana pada seluruh skenario simulasi kapal berhasil mengeksekusi manuver dan menghindari halangan sesuai dengan standar keselamatan maritim global. Selain menjamin keamanan navigasi, penerapan metode kolokasi Chebyshev ini terbukti sangat efisien secara komputasi dengan keberhasilannya dalam mengurangi waktu komputasi secara signifikan sebesar 96,87% dibandingkan dengan penggunaan metode MPC konvensional.
======================================================================================================================================
This research aims to design a heading control system for a SIGMA-class Extended-Corvette warship for dynamic inter-ship collision avoidance in compliance with the international COLREGS using Model Predictive Control (MPC) based on the Chebyshev collocation method. By utilizing a 2 degree of freedo} (sway and yaw) mathematical model and a pseudospectral discretization approach, this algorithm is designed to overcome the high computational burden of conventional MPC through the transcription of the optimal control problem into Quadratic Programming at the Chebyshev-Gauss-Lobatto points. The testing of the Chebyshev MPC algorithm is conducted through simulations using MATLAB, encompassing 4 navigation test scenarios such as a straight-line trajectory without obstacles, overtaking, head-on, and crossing situations. The test results demonstrate that the proposed path planning algorithm operates with high precision. In all simulation scenarios, the ship successfully executes maneuvers and avoids inter-ship collisions in accordance with global maritime safety standards. In addition to ensuring navigational safety, the implementation of this Chebyshev collocation method proves to be highly computationally efficient, yielding a significant computational time reduction of 96.87% compared to the use of conventional MPC methods.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kapal Tanpa Awak, COLREGS, Model Predictive Control, Metode Kolokasi Chebyshev, Penghindaran Halangan, Autonomous Ships, COLREGS, Model Predictive Control, Chebyshev Collocation Method, Obstacle Avoidance
Subjects:	Q Science > QA Mathematics > QA401 Mathematical models.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Zaky Zulfikar
Date Deposited:	23 Jul 2026 06:31
Last Modified:	23 Jul 2026 06:31
URI:	http://repository.its.ac.id/id/eprint/136192

Actions (login required)

View Item