Determining Solution Space Characteristics for Real-Time Strategy Games and Characterizing Winning Strategies

Kurt Weissgerber,Brett J Borghetti,Gary B Lamont,Gilbert L Peterson

doi:10.1155/2011/834026

Kurt Weissgerber, Brett J Borghetti + Show 2 more

Open Access

https://doi.org/10.1155/2011/834026

Copy DOI

Abstract

The underlying goal of a competing agent in a discrete real-time strategy (RTS) game is to defeat an adversary. Strategic agents or participants must define an a priori plan to maneuver their resources in order to destroy the adversary and the adversary's resources as well as secure physical regions of the environment. This a priori plan can be generated by leveraging collected historical knowledge about the environment. This knowledge is then employed in the generation of a classification model for real-time decision-making in the RTS domain. The best way to generate a classification model for a complex problem domain depends on the characteristics of the solution space. An experimental method to determine solution space (search landscape) characteristics is through analysis of historical algorithm performance for solving the specific problem. We select a deterministic search technique and a stochastic search method for a priori classification model generation. These approaches are designed, implemented, and tested for a specific complex RTS game, Bos Wars. Their performance allows us to draw various conclusions about applying a competing agent in complex search landscapes associated with RTS games.

Highlights

The real-time strategy (RTS) domain [1] is of interest because it relates to real world problems, for example, determining a “good” military battlefield strategy or defining the “best” strategies for complex RTS video games
Note that the RTS genre is different than games requiring only real-time tactics (RTT) which deal with making decisions on detailed resource use at each iteration of the game
We address specific RTS game attributes that have a direct consideration in our “optimal” agent algorithmic approach: Case-Based Reasoning, Reinforcement Learning, Dynamic Scripting, and Monte-Carlo planning, along with available RTS software platforms

Summary

Introduction

The real-time strategy (RTS) domain [1] is of interest because it relates to real world problems, for example, determining a “good” military battlefield strategy or defining the “best” strategies for complex RTS video games. A participants/agent strategy is to develop a long-term plan using an agent’s resources to win the game. RTT sometimes are considered a subgenus of real-time strategies. Another way of defining an RTS structure is to consider the terms macro-management referring to high-level strategic maneuvering and micromanagement referring to RTT game interaction. It is desired to gather or destroy resources, build physical structures, improve technological development, and control other agents. This is a daunting set of strategic tasks for an RTS game player

Objectives

Methods

Results