Purba, KristoRadion (2013) Implementasi Algoritma Genetika dan Logika Fuzzy untuk Optimasi Perlengkapan dan Perilaku Musuh dalam Permainan Action-RPG. Magister thesis, Universitas Brawijaya.
Abstract
Dalam penelitian ini, dirancang sebuah permainan bertipe Action-RPG yang berjudul " Song of Ruination 2". Dalam permainan ini diterapkan algoritma genetika untuk mengoptimasi perlengkapan yang dipakai oleh musuh dan pemain, dan juga diterapkan logika fuzzy untuk mengoptimasi perilaku musuh. Dalam permainan ini, setiap musuh dan pemain memiliki status kekuatan, yang terdiri dari attack (Kekuatan serang), defense (Ketahanan), speed (Kecepatan gerak), agility (Kemampuan menghindar) dan life (Nyawa). Nilai dari status ini dapat berubah-ubah dengan dinamis. Algoritma genetika akan mengoptimalkan perlengkapan ( Gloves, armor, shoes ) yang dipakai oleh musuh dan pemain, sehingga status musuh dan pemain merata. Pemerataan status dilihat dari nilai standar deviasi, semakin kecil semakin baik. Selain algoritma genetika, permainan ini juga menerapkan logika fuzzy untuk perilaku musuh. Terdapat 3 jenis musuh, yaitu penyerang, pemanah dan boss . Musuh penyerang mundur ketika nyawanya lemah atau jaraknya dekat dengan pemain. Musuh penembak mundur ketika didekati pemain atau ketika amunisinya sedikit. Musuh boss sangat agresif, ia mundur ketika nyawanya sangat sedikit. Dari hasil penelitian dan pengujian kepada beberapa responden, algoritma genetika ini menghasilkan permainan yang sedikit semakin sulit karena perlengkapan yang dipakai musuh optimal. Logika fuzzy menghasilkan permainan yang secara signifikan semakin sulit, karena perilaku musuh semakin cerdik.
English Abstract
In this study, the writer developed an Action-RPG game entitled "Song of Ruination 2". In this game, genetic algorithms is applied to optimize the equipments used by the enemies and the player, and also fuzzy logic to optimize the behavior of an enemy. In this game, each player has the status of an enemy and force, which consists of attack, defense, speed, agility (ability to dodge) and life. The value of this status can change dynamically. Genetic algorithm will optimize the equipments (gloves, armor, shoes) used by the enemy and the player, so the statuses of the enemy and the player becomes more equal to the status average. Equity of statuses are seen from the standard deviation, the smaller the better. In addition to genetic algorithms, this game also apply fuzzy logic to control the behavior of an enemy. There are 3 types of enemies, the attacker, archers and boss. Attackers retreated when his life is weak or they are close to the players. Shooters retreated when player approached or when their ammunition becomes fewer. Boss is a very aggressive enemy, he retreated when his life very little. From the results of research and testing by several respondents, the genetic algorithm makes the game a little more difficult because of the enemies` equipment is optimized. Fuzzy logic makes the game significantly more difficult, because the behavior of the enemies are smarter.
| Item Type: | Thesis (Magister) |
|---|---|
| Identification Number: | TES/006.31/PUR/i/041307058 |
| Subjects: | 000 Computer science, information and general works > 006 Special computer methods > 006.3 Artificial intelligence |
| Divisions: | S2/S3 > Magister Teknik Elektro, Fakultas Teknik |
| Depositing User: | Endro Setyobudi |
| Date Deposited: | 10 Feb 2014 13:38 |
| Last Modified: | 10 Feb 2014 13:38 |
| URI: | http://repository.ub.ac.id/id/eprint/155424 |
Actions (login required)
 |
View Item |