Title :
Learning of biological behaviour by classifier
Author :
Kanoh, Hideaki ; Hosokawa, Akihiro
Author_Institution :
Sch. of Sci. & Technol., Meiji Univ., Kawasaki, Japan
Abstract :
This paper investigates biological features of a virtual creature on the computer. It is assumed that the creature does not have a complex judgement, a sophisticated detector and has simple basic actions. The classifier system is used for the generation of its behaviour in a complicated environment made in the computer. The results show that the complicated action is generated by a series of simple actions, and that the complex action depends on a complex environment (attraction to the bait, escape from the enemy and avoidance of the obstacle), the aspect of acquisition of the new behaviour pattern is examined in detail
Keywords :
artificial life; evolutionary computation; learning (artificial intelligence); pattern classification; attraction to bait; avoidance of obstacles; behaviour pattern; biological behaviour; bucket brigade; classifier; classifier system; complex action; complex environment; escape from enemy; genetic algorithm; learning; simple basic actions; virtual creature; Animals; Biology computing; Constitution; Detectors; Genetic algorithms; Insects; Research and development; Virtual environment;
Conference_Titel :
Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, International Workshop on, 2001.
Conference_Location :
Crimea
Print_ISBN :
0-7803-7164-X
DOI :
10.1109/IDAACS.2001.942019
