Fuzzy Hunter Optimizer: An Bio-Metaheuristic Algorithm Inspired by L\'evy Walks

Abstract

This article introduces the Fuzzy Hunter Optimizer (FHO), a novel metaheuristic inspired by L\'evy diffuse visibility walk observed in predatory species and even in human behavior during the search for sustenance. To address a constrained optimization problem, we initialize a population of hunters in the search space. The hunter with the best fitness represents the food source. The other hunters move through the search space following a L\'evy walk. When they spot the food source, they move towards it, gradually abandoning the Levy walk. To model the hunters visibility, we employ linear membership functions. In each iteration, the hunter with the best fitness becomes the food source. Unlike other metaheuristics, FHO parameters (visibility functions) do not require pre-calibration, since they adapt with iterations.