A Hierarchical Reinforcement Learning Based Artificial Intelligence for Non-Player Characters in Video Games

Ponce, Hiram; Padilla, Ricardo

doi:10.1007/978-3-319-13650-9_16

A Hierarchical Reinforcement Learning Based Artificial Intelligence for Non-Player Characters in Video Games

Journal

Nature-Inspired Computation and Machine Learning

Lecture Notes in Computer Science

ISSN

0302-9743

1611-3349

Date Issued

2014

Author(s)

Padilla, Ricardo

Type

Resource Types::text::book::book part

DOI

10.1007/978-3-319-13650-9_16

URL

https://scripta.up.edu.mx/handle/20.500.12552/4484

Abstract

Nowadays, video games conforms a huge industry that is always developing new technology. In particular, artificial intelligence techniques have been used broadly in the well-known non-player characters (NPC) given the opportunity to users to feel video games more real. This paper proposes the usage of the MaxQ-Q hierarchical reinforcement learning algorithm in non-player characters in order to increase the experience of the user in terms of naturalness. A case study of an NPC with the proposed artificial intelligence based algorithm in a first personal shooter video game was developed. Experimental results show that this implementation improves naturalness from the user’s point of view. In addition, the proposed MaxQ-Q based algorithm in NPCs allow to programmers a robust way to give artificial intelligence to them.