Observatorio de I+D+i UPM

Memorias de investigación
Sistema Evolutivo Bio-inspirado en el Comportamiento Bacteriano
Research Areas
  • Artificial intelligence (neuronal nets, expert systems, etc)
The main goal of this thesis is to build a new evolutionary computation method that implements an asynchronous, decentralized and continuous general-purpose evolutionary process designed to automatically generate intelligent systems. The grammar-guided evolutionary automatic system (GGEAS) is an evolutionary framework with a modular design that is capable of adapting grammar-guided genetic programming techniques for use in the construction of intelligent systems for different application domains. GGEAS has been used to automatically build symbolic and sub-symbolic intelligent systems, as well as synthetic biological circuits. The artificial bacterium is an evolutionary vehicle bio-inspired by the behavior of bacteria in nature. This bacterium uses a derivation tree belonging to a context-free grammar to codify an intelligent system in its inside. The artificial bacterium constantly and asynchronously evolves this intelligent system as it moves within a constantly changing simulated 3D environment. Additionally, this research defines the conjugation operator and the quorum sensing population control method. The conjugation operator uses a grammar-based crossover operator to cross the inner intelligent systems of two artificial bacteria that are close together in the environment. The quorum sensing control method implements a distributed artificial bacterial population control mechanism. The combined application of the techniques developed in this thesis builds the bacterially inspired evolutionary system. This system takes inspiration from the natural behavior of bacterial populations to build a distributed, asynchronous and parallel evolutionary mechanism that preserves AGGES strengths and is capable of automatically generating intelligent systems in constantly changing environments.
Mark Rating
Sobresaliente cum laude
  • Director: Daniel Manrique Gamo (UPM)
  • Autor: José María Font Fernández (UPM)
Research Group, Departaments and Institutes related
  • Creador: Grupo de Investigación: Grupo de Inteligencia Artificial (LIA)
  • Departamento: Inteligencia Artificial
S2i 2020 Observatorio de investigación @ UPM con la colaboración del Consejo Social UPM
Cofinanciación del MINECO en el marco del Programa INNCIDE 2011 (OTR-2011-0236)
Cofinanciación del MINECO en el marco del Programa INNPACTO (IPT-020000-2010-22)