A Kalman-Filter-Based Sine-Cosine Algorithm

This paper presents a Kalman-Filter-based Sine Cosine algorithm (KFSCA). It is a synergy of a Simulated Kalman Filter (SKF) algorithm and a Sine Cosine (SCA) algorithm. SKF is a random based optimization algorithm inspired from the Kalman Filter theory. A Kalman gain is formulated following the prediction, measurement and estimation steps of the Kalman filter design. The Kalman gain is utilized to introduce a dynamic step size of a search agent in the SKF algorithm. On the other hand, a Sine Cosine algorithm is formulated based on mathematical sine and cosine terms. A random based searching strategy is formulated through a little modification on both of the terms. In the KFSCA, a Kalman gain is introduced to vary an individual agent’s step and thus balances exploration and exploitation strategies of the original SCA. Cost function value that represent an accuracy of a solution is considered as the ultimate goal. Every single agent carries an information about the accuracy of a solution in which will be used to compare with other solutions from other agents. A solution that has a lower cost function is considered as the best solution. The algorithm is tested with various benchmark functions and compared with the original SCA and SKF algorithms. Result of the analysis on the accuracy tested on the benchmark functions is tabulated in a table form and shows that the proposed algorithm outperforms SCA and SKF significantly. The result also is presented in a graphical form to have a clearer visual on the solution.