GA and ACO-based Hybrid Approach for Continuous Optimization

Abstract

This paper presents an hybrid algorithm based on genetic algorithm and ant colony optimization for continuous optimization, which combines the global exploration ability of the former with the local exploiting ability of the later. The proposed algorithm is evaluated on several benchmark functions. The simulation results show that the proposed algorithm performs quite well and outperforms classical ant colony optimization and genetic algorithm for continuous optimization, which efficiently balances two contradictory aspects of its performance: exploration and exploitation.

Ant colony optimization–evolutionary hybrid optimization with translation of problem representation

“…For the purpose of solving continuous optimization problems, ACO–EA hybrids have also been designed. As an example, Reference 32 proposes such an algorithm where ACO R and CGA R (conditionally breeding genetic algorithm) execute their iterations and generations interchangeably while sharing a population. The approach is tested against several benchmark functions and would most likely be classified as an HTH.…”