Pengembangan Gim 3D Tower Defense Menggunakan Peta Procedural Content Generation Berbasis Particle Swarm Optimization

Ramadhan, Mu'aafii (2026) Pengembangan Gim 3D Tower Defense Menggunakan Peta Procedural Content Generation Berbasis Particle Swarm Optimization. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025221135_Undergraduate_Thesis.pdf Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Permainan bergenre Tower Defense (TD) konvensional sering kali memiliki keterbatasan dalam hal replayability akibat penggunaan desain level yang statis dan repetitif. Selain itu, perspektif sudut pandang atas yang umum digunakan cenderung membatasi pemain dalam merasakan atmosfer permainan. Penelitian ini bertujuan untuk mengatasi permasalahan tersebut dengan mengembangkan permainan Tower Defense (TD) yang menggunakan sudut pandang orang ketiga, serta menerapkan metode Procedural Content Generation (PCG) untuk menciptakan variasi level yang dinamis. Algoritma Particle Swarm Optimization (PSO) diterapkan untuk mengoptimalkan tata letak ruang pada peta bertingkat. Setiap partikel dalam algoritma merepresentasikan potensial desain peta yang diiterasi untuk mencapai kriteria kepadatan dan keterhubungan area yang optimal. Untuk menjamin peta yang dihasilkan dapat dimainkan, sistem dilengkapi dengan algoritma pencarian jalur A-Star yang berfungsi memvalidasi konektivitas antar lantai dan melakukan mekanisme pembuatan jalan baru jika terdapat area yang buntu. Pengujian dilakukan menggunakan metode metrik penguiian dalam Game Engine Unity. Hasil pengujian menunjukkan bahwa algoritma PSO mampu menghasilkan peta vertikal tiga lantai yang unik pada setiap sesi permainan dengan tetap mempertahankan struktur yang valid. Penerapan sudut pandang Third-Person berhasil memberikan pengalaman visual yang berbeda dalam strategi penempatan turet, di mana pemain dapat meninjau pertahanan dan berinteraksi dengan lingkungan secara lebih dekat dibandingkan perspektif tradisional.
========================================================================================================================
=========
Conventional Tower Defense (TD) games often suffer from limited replayability due to static and repetitive level designs. Furthermore, the commonly used top-down perspective tends to limit player immersion within the game atmosphere. This research aims to address these issues by developing a hybrid game that combines a Third-Person perspective with Tower Defense strategy elements, applying Procedural Content Generation (PCG) methods to create dynamic level variations.
The Particle Swarm Optimization (PSO) algorithm is applied to optimize the spatial layout of multi-layer maps. Each particle in the algorithm represents a map design candidate that evolves to achieve optimal criteria for area density and connectivity. To ensure the generated maps are playable, the system is equipped with the A-Star pathfinding algorithm to validate inter-floor connectivity and perform automatic path carving mechanisms if dead ends are detected.
Testing was conducted using the Unity Game Engine. The results show that the PSO algorithm is capable of generating unique three-story vertical maps in each game session while maintaining a valid structure. The integration of Third-Person Controller mechanics and defense building systems successfully creates an adaptive gameplay experience, where players are required to devise turret placement strategies while actively participating in defending the base from enemy attacks.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Particle Swarm Optimization (PSO), Procedural Content Generation (PCG), Tower Defense, Third-Person View, Unity 3D. Particle Swarm Optimization (PSO), Procedural Content Generation (PCG), Tower Defense, Third-Person View, Unity 3D.
Subjects:	G Geography. Anthropology. Recreation > GV Recreation Leisure > GV1469.2 Computer games
Divisions:	Faculty of Information Technology > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Mu'aafii Putra Ramadhan
Date Deposited:	27 Jul 2026 03:03
Last Modified:	27 Jul 2026 03:03
URI:	http://repository.its.ac.id/id/eprint/137721

Actions (login required)

View Item