Procedural generation of complex stable structures for angry birds levels

Abstract

This paper presents a procedural content generation algorithm for the physics-based puzzle game Angry Birds. The proposed algorithm creates complex stable structures using a variety of 2D objects. These are generated without the aid of pre-defined substructures or composite elements. The structures created are evaluated based on a fitness function which considers several important structural aspects. The results of this analysis in turn affects the likelihood of particular objects being chosen in future generations. Experiments were conducted on the generated structures in order to evaluate the algorithm's expressivity. The results show that the proposed method can generate a wide variety of 2D structures with different attributes and sizes.