GOHBA: Improved Honey Badger Algorithm for Global Optimization.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• The standard Honey Badger Algorithm (HBA) suffers from local convergence and limited global search capabilities.
• Premature convergence and slow speed hinder the effectiveness of traditional HBA in complex optimization scenarios.

Purpose of the Study:

• To address the limitations of the HBA, this study introduces the Global Optimization Honey Badger Algorithm (GOHBA).
• The GOHBA aims to improve population search, enhance local optimum jumping ability, and increase convergence speed and stability.

Main Methods:

• Initialization using Tent chaotic mapping to improve population diversity and quality.
• Replacement of the density factor to broaden the search range and prevent premature convergence.
• Integration of the golden sine strategy to boost global search capability and accelerate convergence.

Main Results:

• The GOHBA achieved optimal mean values on 14 out of 23 benchmark functions compared to seven other algorithms.
• The GOHBA demonstrated optimal performance on two real-world engineering design problems.
• In path planning problems, the GOHBA exhibited higher accuracy and faster convergence rates.

Conclusions:

• The GOHBA significantly outperforms existing algorithms in terms of global search ability and convergence speed.
• The proposed enhancements effectively mitigate local convergence issues, leading to superior optimization performance.
• The GOHBA proves to be an excellent and robust optimization algorithm for diverse applications.