Efficient Fault Tolerance Methodology in Fanet Using Aco and Ml Techniques

Abstract

An innovative approach is presented in this study to enhance the performance of Ant Colony Optimization (ACO), a type of Bio-Inspired Algorithm (BIA), by integrating machine learning (ML) techniques for fault prediction. The goal is to address the challenges of high end-to-end delay and susceptibility to faults in traditional ACO implementations by leveraging ML methods. Through the application of ML techniques to optimize ACO efficiency and anticipate faults using the Random Forest model, significant reductions in end-to-end delay and improvements in system survivability are achieved. Additionally, the utilization of Least Absolute Shrinkage and Selection Operator (LASSO) feature selection streamlines the optimization process and enhances overall performance. Experimental results demonstrate the superiority of the proposed ML-enhanced ACO approach, indicating its potential for real-world applications in optimization problems.