EVOLUTIONARY LEARNING STRATEGIES FOR ARTIFICIAL LIFE CHARACTERS

摘要

This chapter describes the incorporation of an adaptive learning model to the framework we have designed to implement artificial life characters. This model is based on language structures representing actions and strategies developed by these artificial characters to solve their problems. A set of problems related to how characters evolve and learn may be studied here, ranging from basic survival in their environment to the emergence of knowledge exchange supported by the usage of language to communicate ideas. Furthermore, it presents also a virtual reality application that has been implemented at the USP Digital CAVE. It is an aquarium inhabited by artificial fishes that are able to evolve in this environment using their learning ability. These fishes have their own cognition, which control their actions - mainly swimming and eating. Through contact and communication with other fishes they learn how to behave in the aquarium, trying to remain alive. The simulation has been implemented in JAVA 3D. A main server and five clients compose the distributed VR version. The server comprehends the simulation of the aquarium and contained fishes, while the clients are responsible for the different views of this aquarium (from inside) projected at each of the five CAVE walls.