Aditama, Darmawan (2016) Evolusi Dinamis Perilaku Non-Player Character Pada Game Space Shooter Menggunakan NSGA-II. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2214205013-Master_Theses.pdf - Published Version Download (3MB) | Preview |
Abstract
Dalam permainan space shooter terdapat musuh yang dikendalikan oleh Non-Player Character (NPC) statis dimana player dapat terus berusaha beradaptasi terhadap perilaku dari musuh. Pada akhirnya, permainan menjadi membosankan karena perilaku NPC telah diketahui oleh player. Dalam penelitian ini dikembangkan implementasi kecerdasan buatan dimana NPC dapat mengevolusi dirinya sendiri, sehingga mampu merespon perilaku player. Evolusi dinamis NPC dapat memberikan pengalaman bermain yang menyenangkan karena perilaku NPC dapat menyesuaikan perilaku player. Non-Dominated Sorting Genetic Algorithm II (NSGA-II) dalam simulasi digunakan untuk mengatur “titik evolusi” terhadap NPC. Dimana dengan memeriksa parameter keputusan dari perilaku playar untuk dapat mengevolusi perilaku dari musuh (NPC). Kemudian Non-Dominated Sorting Genetic Algorithm II (NSGA-II) digunakan untuk mencari optimal solution untuk mengoptimalkan evolusi dinamis NPC pada level berikutnya. Hal ini bertujuan untuk menyeimbangkan perilaku musuh terhadap player. Simulasi menggunkan NSGA-II menghasilkan solusi optimal dalam bentuk grafik dengan 2 obyektif (speed dan health) dari NPC. Berdasarkan hasil simulasi, kestabilan berada pada generasi ke-5 dengan populasi=50, probabilitas persilangan (pc) = 1, probabilitas mutasi (pm)= 1/n, indeks distribusi persilangan (η_c)=20, dan indeks distribusi mutasi (η_m)=20.
==================================================================================================================
Abstrak – In Space Game, the NPC enemies have a static ability or behaviour where the player can be easily learn to adapted and challenge the NPCs. In the end, it makes the player become boring and don't want to play the game anymore. This research is develop an intelligent NPC that can be adapt to player style in playing the game. The NPC can be evolve and adjust itself based on the player behaviour in playing game and giving a proper difficulty to the game itself. This implementation will be make game more fun and enjoyable. NSGA II is the algorithm that will used to arrange the evolution of NPCs. This algorithm is used to evaluate the parameter that will determine NPCs behaviour change. Beside that, this research also used NSGA-II to optimized the paramater before it used to determined the NPCs behaviour change in every level of the game. Simulation it self used NSGA-II to produce optimal solution in graphic representation based on two objective parameter, speed and health. Result of simulation shown the optimized result will be produce the best solution after the five generation with total population is 50. crossover probability (pc)=0.9, mutation probability (pm)=1/n, index of distribution crossover (ηc)=20, index of distribution mutation (ηm) =20.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Genetics Algorithm, Artificial Intelligent, Behavior Evolution |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | - DARMAWAN ADITAMA |
| Date Deposited: | 16 Mar 2017 02:31 |
| Last Modified: | 26 Dec 2018 03:41 |
| URI: | http://repository.its.ac.id/id/eprint/1904 |
Actions (login required)
 |
View Item |