Proposed Solution to Overcome the Shortcomings of Glow Worm Swarm Optimization Algorithm
Sannaullah, Department of Computing and Technology, Abasyn University, Peshawar, Pakistan.
Syed Aizaz Ul Haq, Department of Computing and Technology, Abasyn University, Peshawar, Pakistan.
Abdul Manan Ahmadzai, Department of Computing and Technology, Abasyn University, Peshawar, Pakistan.
Muhammad Asif Sahim, Department of Computing and Technology, Abasyn University, Peshawar, Pakistan.
Corresponding Author:
Sannaullah (sannaullah6464@gmail.com)
Abstract:
Swarm intelligence, which is inspired by the group behavior of social insects and other animals, serves as the foundation for this study. A swarm is a collection of fish, birds, ants, bees, etc. Animal agents are employed in this swarm to maximize the issue. For large-scale optimization issues, where traditional mathematical techniques would not work, optimization algorithms have been industrialized to provide near-optimal solutions. Algorithms for optimization are employed to minimize or maximize an objective function. Without any centralized information, each agent in the swarm operates independently, sharing knowledge and progressing together. Nevertheless, optimization techniques have their own set of drawbacks, which are mainly related to the restrictions of the objects they aim to simulate. Even though GSO is commonly utilized, it often experiences early convergence and may not have enough diversity in the glowworm population, which could hinder its capacity to travel away from local optima. The current project aims to address the GSO's shortcomings.
Keywords:
Optimization Algorithm; Meta-Heuristic; Glow Worm Swarm (GSO); Particle Swarm Optimization (PSO); Grey Wolf Optimization (GWO); Convergence Rate.