Optimisasi Parameter IFDS Menggunakan Genetic Algorithm Untuk Perencanaan Jalur Multi-Rintangan Pada Quadcopter

Wahyunirmala, Afchrysillva Zahrani (2026) Optimisasi Parameter IFDS Menggunakan Genetic Algorithm Untuk Perencanaan Jalur Multi-Rintangan Pada Quadcopter. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221013-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Sistem navigasi UAV, khususnya quadcopter, memerlukan metode perencanaan jalur yang mampu menghasilkan lintasan yang efisien, halus untuk diikuti, serta aman terhadap rintangan pada lingkungan yang kompleks. Pada penelitian ini dikembangkan sebuah simulasi untuk perencanaan jalur quadcopter pada skenario multi-rintangan dengan dua tahap perencanaan, yaitu perencanaan jalur global menggunakan Interfered Fluid Dynamical System (IFDS) dan perencanaan jalur lokal untuk rintangan dinamis menggunakan Fuzzy-IFDS dengan mekanisme switching. Pemilihan parameter IFDS yang berpengaruh terhadap kualitas jalur dilakukan menggunakan Genetic Algorithm (GA) untuk mengoptimalkan parameter ρ (koefisien repulsi), σ (koefisien tangensial), θ (koefisien sudut), λ (kecepatan relatif) dioptimalkan menggunakan GA untuk meminimalkan cost function, dengan tetap mengevaluasi metrik path length, mean curvature, minimum clearance, serta waktu komputasi. Pengujian pengaruh parameter GA dilakukan dengan memvariasikan jumlah populasi 〖(N〗_pop) dan batas generasi maksimum (MaxGen). Hasil pengujian menunjukkan bahwa konfigurasi terbaik pada variasi jumlah populasi diperoleh saat N_pop=20, dengan penurunan cost sebesar 0,63%, penurunan path length sebesar 1,79%, serta mean curvature sebesar 2,5% dibandingkan konfigurasi awal N_pop=10. Pada variasi generasi maksimum diperoleh saat MaxGen=30 dengan nilai cost turun 0,6% dibanding MaxGen=10 serta peniingkatan minimum distance sebesar 5,03%. Untuk rintangan dinamis, Fuzzy-IFDS secara online menyesuaikan parameter ρ dan θ secara adaptif ketika UAV memasuki area switching berdasarkan informasi jarak dan sudut relatif rintangan, kemudian mengembalikan parameter menuju nilai baseline setelah rintangan terlewati sehingga UAV dapat melanjutkan pergerakan menuju target. Dengan demikian, kombinasi GA dan Fuzzy-IFDS mampu meningkatkan kualitas lintasan global serta mempertahankan keselamatan navigasi pada skenario multi-rintangan.
=====================================================================================================================================
A UAV navigation system, particularly for a quadcopter, requires a path-planning method that can produce trajectories that are efficient, smooth to follow, and safe with respect to obstacles in complex environments. In this study, develop based simulation is developed for quadcopter path planning in a multi-obstacle scenario using a two-stage planning framework: global path planning based on the Interfered Fluid Dynamical System (IFDS) and local planning for dynamic obstacles using a Fuzzy-IFDS approach with a switching mechanism. The selection of IFDS parameters that influence trajectory quality is carried out using a Genetic Algorithm (GA), in which the parameters ρ (repulse coefficient), σ (tangensial coefficient), θ (angular coefficient), λ (relative speed) are optimized to minimize a cost function, while also evaluating the path length, mean curvature, minimum clearance, and computational time. The effects of GA parameters are examined by varying the population size 〖(N〗_pop) and the maximum number of generations (MaxGen). The results show that the best configuration for population size is obtained at N_pop=20, achieving a cost reduction of 0,63%, a path length reduction of 1,79%, and a mean curvature reduction of 2,5% compared to the initial configuration with N_pop=10. For the maximum generation variation, the best performance is achieved at MaxGen = 30, where the cost decreases by 0,6% compared to MaxGen = 10 and the minimum distance increases by 5,03%. For dynamic obstacles, the Fuzzy-IFDS module adaptively adjusts the parameters ρ and θ online when the UAV enters the switching region, based on obstacle distance and relative angle information, and then returns the parameters toward their baseline values after the obstacle is cleared so that the UAV can continue toward the target. Thus, the combination of GA and Fuzzy-IFDS improves the quality of the global trajectory while maintaining safe navigation in multi-obstacle scenarios.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Quadcopter, Genetic Algorithm, IFDS, Fuzzy, Path planning, Quadcopter, Genetic Algorithm, IFDS, Fuzzy, Path planning
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ211 Robotics. T Technology > TJ Mechanical engineering and machinery > TJ217 Adaptive control systems T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control 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
Divisions:	Faculty of Electrical Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Afchrysillva Zahrani Wahyunirmala
Date Deposited:	27 Jan 2026 09:00
Last Modified:	27 Jan 2026 09:00
URI:	http://repository.its.ac.id/id/eprint/130550

Actions (login required)

View Item