TY - JOUR
T1 - IORand
T2 - A Procedural Videogame Level Generator Based on a Hybrid PCG Algorithm
AU - Moreno-Armendáriz, Marco A.
AU - Calvo, Hiram
AU - Torres-León, José A.
AU - Duchanoy, Carlos A.
N1 - Publisher Copyright:
© 2022 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2022/4/1
Y1 - 2022/4/1
N2 - In this work we present the intelligent orchestrator of random generators (IORand), a hybrid procedural content generation (PCG) algorithm, driven by game experience, based on reinforcement learning and semi-random content generation methods. Our study includes a presentation of current PCG techniques and why a hybridization of approaches has become a new trend with promising results in the area. Moreover, the design of a new method for evaluating video game levels is presented, aimed at evaluating game experiences, based on graphs, which allows identifying the type of interaction that the player will have with the level. Then, the design of our hybrid PCG algorithm, IORand, whose reward function is based on the proposed level evaluation method, is presented. Finally, a study was conducted on the performance of our algorithm to generate levels of three different game experiences, from which we demonstrate the ability of IORand to satisfactorily and consistently solve the generation of levels that provide specific game experiences.
AB - In this work we present the intelligent orchestrator of random generators (IORand), a hybrid procedural content generation (PCG) algorithm, driven by game experience, based on reinforcement learning and semi-random content generation methods. Our study includes a presentation of current PCG techniques and why a hybridization of approaches has become a new trend with promising results in the area. Moreover, the design of a new method for evaluating video game levels is presented, aimed at evaluating game experiences, based on graphs, which allows identifying the type of interaction that the player will have with the level. Then, the design of our hybrid PCG algorithm, IORand, whose reward function is based on the proposed level evaluation method, is presented. Finally, a study was conducted on the performance of our algorithm to generate levels of three different game experiences, from which we demonstrate the ability of IORand to satisfactorily and consistently solve the generation of levels that provide specific game experiences.
KW - artificial intelligence
KW - hybrid algorithms
KW - procedural content generation
KW - reinforcement learning
KW - semi-random generation
UR - http://www.scopus.com/inward/record.url?scp=85128716121&partnerID=8YFLogxK
U2 - 10.3390/app12083792
DO - 10.3390/app12083792
M3 - Artículo
AN - SCOPUS:85128716121
SN - 2076-3417
VL - 12
JO - Applied Sciences (Switzerland)
JF - Applied Sciences (Switzerland)
IS - 8
M1 - 3792
ER -