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

TL;DR

The paper introduces the Fuzzy Hunter Optimizer, a new bio-inspired metaheuristic algorithm based on Le9vy walks and fuzzy visibility modeling, which adaptively searches for optimal solutions without pre-calibrated parameters.

Contribution

It presents a novel metaheuristic inspired by predatory behavior and Le9vy walks, with adaptive visibility functions that do not require pre-calibration.

Findings

Demonstrates effectiveness on constrained optimization problems.

Features adaptive parameters that evolve during the search process.

Outperforms some existing metaheuristics in benchmark tests.

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.