Inteligencia computacional distribuida : arquitetura especificação formal e aplicação

AUTOR(ES)
DATA DE PUBLICAÇÃO

1996

RESUMO

Distributed artificial intelligence aims at creating flexible models for problem solving through intelligent agents. Agents have to interact with one another seeking for cooperation, despite their incomplete, uncertain or even inconsistent knowledge. Fuzzy sets theory has been a focus of intense research in the last years, both in the theoretical as well as in the application fields. Its use has spread over for its mechanisms for handling uncertainties and for the flexibility it provides. Case-based systems provides mechanisms to mimic human behavior in decision making through the use of past decisions models in similar situations. They allow complex decision making, re-using previous computational effort. Genetic algorithms are often used for problem solving and decision making in new environments. They lead to adaptation and performance increase, avoiding local maxima or minima. This work constitutes a symbiosis among distributed artificial intelligence, fuzzy sets theory, case-based systems, and genetic algorithms. It aims at introducing an approach that provides distributed problem solving, cooperation, uncertainties handling, flexibility, use of past experiences, and adaptation to new situations. The Distributed Computational lntelligence approach is a basis for the development of a distributed traific control system. Several performance tests are presented, and their results are compared to conventional traific control strategies

ASSUNTO(S)

inteligencia artificial engenharia de sistemas

Documentos Relacionados