Automatic Generation of Real-Time Animation Game Learning Levels Based on Artificial Intelligence Assistant

Abstract

With the development of computer animation games, now many animation games use the automatic generation of animation game levels method to generate animation game content. Using the automatic generation levels of the animation game method can quickly generate many different levels, saving labor costs in manual production. And with the combination of other fields, more accurate results can be obtained. In this paper, the goal is to automatically generate the level of the animation game. This research proposed animation game objects and victory conditions to derive the concept of the clearance path and uses artificial intelligence genetic algorithms to obtain the clearance path that best meets the needs of the developer to recombine new levels. In addition, since the generated levels are regenerated from the clearance path, there are still too many blank areas. In order to make full use of these blank areas, this research provides a level complete system. The system has formulated its placement rules for existing animation game objects. These rules fill in these blank areas and produce a more complete level. The contributions of this search proved that the genetic algorithm can converge quickly due to the correction effect in the mutation. In the complete system, it has successfully achieved a complete and playable level. In the system running time, this research adjusted the method of selecting elites and the use of permutations and combinations to mate, so that the system’s operation can be stabilized on an average value.