A review of animal behavior-inspired methods for intelligent systems

Abstract

This paper examines the most significant approaches towards developing intelligent systems that were inspired by the behavior of natural animals. An overview is presented that describes the essential aspects of several of those methodologies and summarizes the most noteworthy related publications. Emergent behaviors are those behaviors that arise out of complex system and cannot be explained by the specific rules of the system. Emergent behavior is observed from two aspects: those emerging within a single-agent and those occurring in multi-agent systems. Popular algorithms inspired by bats, krills, and bees behavior are also reviewed. Their recent applications in solving optimization, data mining, scheduling, image processing, and similar difficult problems are listed.