EvolvingBehavior: Towards Co-Creative Evolution of Behavior Trees for Game NPCs
Neural and Evolutionary Computing
2022-09-05 v1 Artificial Intelligence
Human-Computer Interaction
Abstract
To assist game developers in crafting game NPCs, we present EvolvingBehavior, a novel tool for genetic programming to evolve behavior trees in Unreal Engine 4. In an initial evaluation, we compare evolved behavior to hand-crafted trees designed by our researchers, and to randomly-grown trees, in a 3D survival game. We find that EvolvingBehavior is capable of producing behavior approaching the designer's goals in this context. Finally, we discuss implications and future avenues of exploration for co-creative game AI design tools, as well as challenges and difficulties in behavior tree evolution.
Keywords
Cite
@article{arxiv.2209.01020,
title = {EvolvingBehavior: Towards Co-Creative Evolution of Behavior Trees for Game NPCs},
author = {Nathan Partlan and Luis Soto and Jim Howe and Sarthak Shrivastava and Magy Seif El-Nasr and Stacy Marsella},
journal= {arXiv preprint arXiv:2209.01020},
year = {2022}
}
Comments
13 pages, 5 figures. Accepted for publication in Foundations of Digital Games 2022 (FDG '22)