The virtual worlds of computer games and similar animated simulations may be populated by autonomous characters that intelligently navigate in virtual cities. We concretely apply hybrid system theory and tools to model navigation strategies for virtual characters. In particular, we present hybrid systems for both low-level (local) and high-level (global) navigation strategies, and we describe how we modeled these systems using the hybrid system specification tool Charon. Further, we directly employed our hybrid system models to generate animations that demonstrate these navigation strategies. Overall, our results suggest that hybrid systems may be a natural framework for modeling aspects of intelligent virtual actors. We also present a small verification example for a simple navigation strategy, and we briefly discuss obstacles to widespread practical applicability of verification in this problem domain.
CITATION STYLE
Aaron, E., Ivančić, F., & Metaxas, D. (2002). Hybrid system models of navigation strategies for games and animations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2289, pp. 7–20). Springer Verlag. https://doi.org/10.1007/3-540-45873-5_4
Mendeley helps you to discover research relevant for your work.