An Oppositional Learning Prediction Operator for Simulated Kalman Filter

Abstract

Simulated Kalman filter (SKF) is a recent metaheuristic optimization algorithm established in 2015. In the present study, we introduce a prediction operator in SKF to prolong its exploration and to avoid premature convergence. The proposed prediction operator is based on oppositional learning. The results show that using CEC2014 as benchmark problems, the SKF algorithm with oppositional learning prediction operator outperforms the original SKF algorithm in most cases.