Pemetaan Dan Penghindaran Rintangan Dinamis Dalam Ruangan Pada TurtleBot 4 Menggunakan SLAM Dan Modifikasi APF

Putranta, Mahesa Bima (2026) Pemetaan Dan Penghindaran Rintangan Dinamis Dalam Ruangan Pada TurtleBot 4 Menggunakan SLAM Dan Modifikasi APF. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221228-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Navigasi otonom mobile robot pada lingkungan indoor dengan rintangan dinamis memerlukan sistem yang mampu melakukan pemetaan, perencanaan jalur, pelacakan jalur, serta penghindaran rintangan secara adaptif. Artificial Potential Field (APF) umum digunakan untuk penghindaran rintangan, namun rentan terhadap permasalahan local minima, khususnya pada area sempit. Penelitian ini mengusulkan penghindaran rintangan dinamis melalui modifikasi APF berbasis sektor. Kinerja sistem dievaluasi berdasarkan jarak minimum robot terhadap rintangan serta efisiensi navigasi yang meliputi durasi tempuh, panjang lintasan, dan akurasi pelacakan jalur. Hasil pengujian menunjukkan bahwa jarak minimum robot terhadap rintangan berada pada rentang 0.17–0.32 m tanpa terjadi tabrakan. Variasi parameter terbaik diperoleh pada konfigurasi Modifikasi APF dengan nilai K_attr = 0.5, K_rep = 0.3, serta radius aktivasi sektor depan 0.7 m, samping 0.3 m, dan belakang 0.25 m, yang menghasilkan durasi tempuh 84.03 s, panjang lintasan 26.006 m, dan nilai RMSE pelacakan sebesar 0.1778 m. Meskipun terjadi peningkatan durasi tempuh dan panjang lintasan dibandingkan pelacakan jalur tanpa penghindaran rintangan, peningkatan tersebut hanya terjadi selama fase aktivasi penghindaran rintangan dinamis, dan berkurang setelah robot kembali mengikuti jalur global melalui kontrol pelacakan jalur, sehingga robot tetap mampu kembali mengikuti jalur global tanpa penyimpangan lintasan yang berlebihan. Modifikasi APF berbasis sektor menghasilkan gaya tolak yang lebih terarah, sehingga mengurangi kecenderungan terjadinya local minima dan memungkinkan robot keluar dari area sempit untuk melanjutkan navigasi menuju tujuan.
=================================================================================================================================
Autonomous navigation of mobile robots in indoor environments with dynamic obstacles requires a system capable of environment mapping, path planning, path tracking, and adaptive obstacle avoidance. Artificial Potential Field (APF) is commonly used for obstacle avoidance; however, it is prone to the local minima problem, particularly in narrow environments. This research proposes a sector-based Modified APF for dynamic obstacle avoidance. System performance is evaluated based on the minimum distance between the robot and obstacles as well as navigation efficiency, including travel time, path length, and path tracking accuracy. Experimental results show that the minimum distance between the robot and obstacles ranges from 0.17 to 0.32 m without any collisions. The best parameter configuration is obtained using the Modified APF with K_attr = 0.5, K_rep = 0.3, and activation radius of 0.7 m for the front sector, 0.3 m for the side sector, and 0.25 m for the rear sector, resulting in a travel time of 84.03 s, a path length of 26.006 m, and a path tracking RMSE of 0.1778 m. Although the travel time and path length increase compared to path tracking without obstacle avoidance, this increase occurs only during the activation phase of dynamic obstacle avoidance and decreases after the robot returns to the global path through the path tracking controller. As a result, the robot is able to resume following the global path without excessive trajectory deviation. The sector-based Modified APF produces more directed repulsive forces, thereby reducing the tendency to fall into local minima and enabling the robot to escape confined areas and continue navigating toward the goal.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Modifikasi APF, Navigasi Otonom, ROS2, SLAM Toolbox, TurtleBot 4, Modified APF, Autonomous Navigation, ROS2, SLAM Toolbox, TurtleBot 4
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ211.415 Mobile robots T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.8 Vehicles, Remotely piloted. Autonomous vehicles.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Mahesa Bima Putranta
Date Deposited:	21 Jan 2026 07:14
Last Modified:	21 Jan 2026 07:14
URI:	http://repository.its.ac.id/id/eprint/129968

Actions (login required)

View Item