Next January at the 7th International Conference on Agents and Artificial Intelligence (ICAART 2015) in Lisbon, Juliette Lemaitre will present her paper “Towards a resource-based model of strategy to help designing opponent AI in RTS games”. Juliette is currently a PhD student at MASA Group and the Heudiasyc lab. She’s working on strategic behaviors in RTS games.
The artificial intelligence used for opponent non-player characters in commercial real-time strategy games is often criticized by players. It is used to discover the game but soon becomes too easy and too predictable. Yet, a lot of research has been done on the subject, and successful complex behaviors have been created, but the systems used are too complicated to be used by the video games industry, as they would need time for the game designer to learn how they function, which ultimately proves prohibitive. Moreover these systems often lack control for the game designer to be adapted to the desired behavior.
This paper addresses the issue by proposing an accessible behavior model that defines a strategy as the decision-making process of the allocation of available resources, such as agents or objects, to sub-tasks in the pursuit of an overall goal suited to the environment, in particular other agents. The strategy consists of a hierarchy of logical and parallel behaviors. The first type allows the game designer to choose an appropriate behavior according to the situation, the second one to allocate the work to the available resources.
This model is part of a bigger system which aims to facilitate the creation of behaviors thanks to a behavior generator, providing the game designer a first strategy to manipulate and improve. It also intends to update the strategy with new game mechanics while keeping it consistent with the previous version, so that it fits into the creation process of a game during which several modifications often occur.
The article is now available here.