Simulasi Pergerakan Pengunjung Mall Menggunakan Potential Field

Kurniawati, Arik (2010) Simulasi Pergerakan Pengunjung Mall Menggunakan Potential Field. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 2208205743-Master_thesis.pdf Restricted to Repository staff only Download (25MB)

Abstract

Simulasi kerumunan manusia secara real-time termasuk sesuatu yang sulit karena perilaku yang ditunjukkan oleh kelompok besar ini sangat kompleks dan rumit. Beberapa hal yang perlu dicermati antara lain penanganan terhadap permasalahan path planning (perencanaan jalur), collision avoidance (penghindaran tabrakan), serta separation (pemisahan jarak). Salah satu metode untuk perencanaan jalur adalah potential field yang berbasis grid. Metode ini memiliki prinsip seperti medan magnet yang menarik atau menolak partikel besi yang ada di sekitarnya. Berdasarkan karakteristik metode ini, maka metode ini dapat digunakan untuk mensimulasikan pergerakan simulasi kerumunan manusia. Perilaku menolak partikel besi diaplikasikan untuk menghindari halangan dinding dan perilaku menarik diaplikasikan untuk keluar menuju pintu utama. Selain itu, juga diatur agar pergerakan antarorang tidak menimbulkan tabrakan. Setelah diadakan simulasi dan penelitian, maka pergerakan kerumunan manusia yang tersebar dalam ruang-ruang dan lorong-lorong di mal dengan menggunakan metode potential field dapat menghindari halangan dinding dan dapat bergerak keluar menuju pintu utama tanpa terjebak local minima. Namun masih ada kendala, yakni bertabrakan dengan orang lain jika daerah yang dituju sama. Tingkat frekuensi tabrakan semakin meningkat seiring meningkatnya jumlah populasi; untuk jumlah populasi 10 terjadi 1 kali tabrakan dan bisa mencapai 9.558 tabrakan untuk jumlah populasi 500.
==================================================================================================================================
Real-time crowd simulation is complicated because large groups of people exhibit behavior of enormous complexity and subtlety. In crowd simulation, there are many factors that must be seriously addressed, such as path planning, collision avoidance, and separation. In this topic, we use potential field for path planning, which is a method based on a grid. The rule of this method is like a magnetic field that attracts or repulses iron particles. Based on this characteristic, we assumed that this method can be implemented to generate crowd simulation movement. Repulsing iron particle behavior can be used to avoid static obstacles and attracting behavior can be used to drive the crowd to the target. Before moving to the next position, each person must process collision prediction to avoid collisions with others. After executing simulations using the potential field method, we determined that the movement of a human crowd scattered in the rooms and hallways of a mall can avoid obstacles (walls) and successfully reach the target without getting trapped in local minima. However, there is still a constraint, namely colliding with other people when heading toward the same destination. The frequency level of collisions increases along with the growing population; for a population of 10, 1 collision occurs, and it can reach up to 9,558 collisions for a population of 500.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	path planning, collision avoidance, separation, potential field, local minima.
Subjects:	T Technology > T Technology (General) > T57.62 Simulation
Divisions:	Faculty of Electrical Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	magang .
Date Deposited:	25 Sep 2026 08:36
Last Modified:	25 Sep 2026 08:36
URI:	http://repository.its.ac.id/id/eprint/144912

Actions (login required)

View Item