ANALYSIS OF BOID ALGORITHM AS A METHOD FOR ANIMAL FLOCKING SIMULATION USING ALIGNMENT CLUSTERING INDEX

• Main Author: Aulia Rahman, Faiz
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/77479
• Institution: Institut Teknologi Bandung

Collective animal behavior exhibits intersting phenomenon found in nature. As they However these phenomenon can be complicated and challenging to simulate and visualize. In 1987 Craig Reynolds introduced an algorithm called boid or bird-oid which simulates bird-like flocking behavior. Comprising specific rules, the algorithm captures individual motion within a flock, later categorized into three components by its creator in 1999 that is: cohesion, alignment, and separation. These rules encourage agents in the simulation to interact and move collectively. This research aims to analyze boid using its simulation. In order to modify how each rule influence the boid movement, a weight value is assigned. These values are varied to obtain the data needed for analysis. Futhermore, additional rules such as bound, containment, and wander are incorporated to the simulation. The simulations are executed using the C#-based Unity engine, version 2020.3.3f1 LTS. The analysis is performed using the alignment clustering index (ACI) due to periodic flocking movement within the same area. Results reveal that alignment is the most influential factor in collective behavior, as evidenced by the increasing average ACI values corresponding to higher alignment weights.