Optimasi Jalur Lintasan Lengan Robot Untuk Menggambar Berbasis Algoritma Genetika

Nurkholik, Zen (2025) Optimasi Jalur Lintasan Lengan Robot Untuk Menggambar Berbasis Algoritma Genetika. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022222007-Master_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Penelitian ini membahas tentang optimasi jalur lintasan lengan robot dengan 3 derajat kebebasan (3-DOF) untuk membuat jalur lintasan berdasarkan sketsa wajah berbentuk titik hasil dari proses deteksi tepi. Citra wajah diolah menggunakan filter Gaussian dan metode deteksi tepi Canny, kemudian dikonversi menjadi kumpulan titik-titik koordinat yang digunakan sebagai dasar pergerakan lengan robot. Optimasi lintasan dilakukan menggunakan tiga algoritma metaheuristik, yaitu Genetic Algorithm (GA), Particle Swarm Optimization (PSO), dan Simulated Annealing (SA). Evaluasi performa dilakukan berdasarkan total jarak lintasan, waktu eksekusi, kecepatan rata-rata, serta efisiensi jumlah generasi/iterasi dan jumlah evaluasi per detik. Hasil eksperimen menunjukkan bahwa GA menghasilkan lintasan terpendek sebesar 3.456,75 unit, dengan kecepatan rata-rata 159,32 unit/detik dan waktu eksekusi selama 21,07 detik. PSO menghasilkan lintasan sepanjang 23.951,77 unit dengan kecepatan 131,72 unit/detik dan waktu eksekusi 181,76 detik. Sementara itu, SA menunjukkan waktu eksekusi tercepat yaitu 0,07 detik dan kecepatan rata-rata sangat tinggi sebesar 355.547,29 unit/detik, namun menghasilkan lintasan paling panjang sebesar 24.888,31 unit. Secara keseluruhan, GA dinilai paling optimal dalam menghasilkan jalur yang efisien dari segi jarak tempuh terbaik. Hasil penelitian ini menunjukkan bahwa optimasi, khususnya dengan pendekatan berbasis GA, dapat dimanfaatkan secara efektif dalam perencanaan jalur lintasan lengan robot yang mengikuti pola sketsa titik dari citra wajah.
=======================================================================================================================================
This research discusses path optimization for a 3 degree of freedom (3-DOF) robotic arm to generate motion paths based on facial sketch points derived from edge detection. The facial image is processed using a Gaussian filter and Canny edge detection, then converted into a set of coordinate points used as the basis for the robot arm's movement. Path optimization is carried out using three metaheuristic algorithms: Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA). The performance evaluation is based on total path distance, execution time, average speed, as well as the efficiency of generations/iterations and evaluations per second. Experimental results show that GA produces the shortest path of 3,456.75 units, with an average speed of 159.32 units/second and an execution time of 21.07 seconds. PSO produces a path length of 23,951.77 units, with a speed of 131.72 units/second and an execution time of 181.76 seconds. Meanwhile, SA shows the fastest execution time of 0.07 seconds and the highest average speed of 355,547.29 units/second, but results in the longest path of 24,888.31 units.Overall, GA is considered the most optimal in producing efficient paths in terms of shortest distance. These results indicate that optimization, particularly with a GA-based approach, can be effectively utilized for robotic arm path planning that follows point based sketches derived from facial image processing.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Optimasi Jalur Lintasan, Lengan Robot, Deteksi Tepi Canny, Path Optimization, Robotic Arm, Canny Edge Detection
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Zen Nurkholik
Date Deposited:	29 Jul 2025 02:02
Last Modified:	29 Jul 2025 02:02
URI:	http://repository.its.ac.id/id/eprint/122651

Actions (login required)

View Item