This work proposes a simple cognitive architecture that allows for simulation of living beings (such as game NPCs) in a variety of different contexts and detail levels. In this repository, specifically, the architecture is used to create a small ecosystem, where virtual animals have their own desires and needs, and are aware of their surroundings.
The current high-level goals are:
- to provide further simple yet representative implementations of cognitive processes of animals (e.g. the fight-or-flight mecanism, working memory, learning, etc).
- to keep the architecture design generic, modular and scalable enough for it to be easily understood, used in other projects and even augmented or adapted.
This was part of my final project for the Computer Science BSc degree at UFRRJ, made under orientation of DSc Bruno Dembogurski.
- Vieira, Ronaldo, et al. "A Cognitive Architecture for Agent-Based Artificial Life Simulation." International Conference on Computational Science and Its Applications. Springer, Cham, 2018.