Strategy games such as Warcraft or UFO franchises or RPG games like Never Winter Nights or Baldur Gate are successful blockbusters in video game industry. These games are based on battles simulated individual per individual. These type of games is a very interesting scenario to develop multilevel strategies or emergent behavior in multiagent systems. This paper presents a new computational intelligence framework, named VBATTLE, for the evaluation of learning strategies in video games. The framework simulates a battle game in which two or more contenders are fighting in units with a high-detail individual-per-individual resolution. This simulation considers aspects of (1) actions parameters (action time, exhaustion consumption), (2) non-deterministic action resolution, (3) hierarchical intelligence (individual vs. unit strategies), and (4) scenario interaction. The VBATTLE framework is designed to have both a 3D visual representation of executions (either on-line or post-mortem visualization) as well as a server-based engine to perform learning tasks. This contribution presents the design principles of VBATTLE framework, its objectives, the possible applications for developing computational intelligence algorithms. In addition, preliminary results using a limited version of the framework are also presented.
展开▼
