A FIRST LOOK AT BUILD-ORDER OPTIMIZATION IN REAL-TIME STRATEGY GAMES

real-time strategy games, concurrent actions Planning is an important cognitive process. We are constantly involved in planning even when executing seemingly simple tasks such as driving to school. The ability to plan becomes essential in unfamiliar environments where we cannot rely on previously learned action sequences, but instead have to generate novel solutions by considering consequences of action choices on the fly. Planning in real-world domains poses a big challenge to current AI systems because of inadequate abstraction and search mechanisms. Human planning ability is therefore still considered superior. A good example is modern video games in which it is apparent that computer characters lack human planning, learning, and reasoning abilities. In this paper we approach the real-time planning problem by considering build-order optimization in real-time strategy games. This problem class can be formulated as a resource accumulation and allocation problem where an agent has to decide what objects to produce at what time in order to meet one of two goals: either maximizing the number of produced objects in a given time period or producing a certain number of objects as fast as possible. We identify challenges related to this planning problem, namely creating and destroying objects, concurrent action generation and execution, and present ideas how to address them