ABSTRACT
Fully scripted game AI systems are usually predictable and, due to statically defined behavior, susceptible to poor decision-making when facing unexpected opponent actions. In games with a small number of possible actions, like chess or checkers, a successful approach to overcome these issues is to use look-ahead search, that is, simulating the effects of action sequences and choosing those that maximize the agent's utility. This chapter presents an approach that adapts this process to complex video games, reducing action choices by means of scripts that expose choice points to look-ahead search. The chapter represents the scripts as decision trees because of the natural formulation of choice points as decision nodes. It uses the Negamax version of the minimax algorithm for simplicity. The chapter introduces scripts with choice points, a state evaluation function, and a search algorithm that uses look-ahead to decide which choices to take.
