Abstract
Inspired by Kalman filtering, simulated Kalman filter (SKF) has been introduced as a new population-based optimization algorithm. The SKF is not a parameter-less algorithm. Three parameter values should be assigned to P, Q, and R, which denotes error covariance, process noise, and measurement noise, respectively. While analysis of P has been studied, this paper emphasizes on Q and R parameters. Instead of using constant values for Q and R, random values are used in this study. Experimental result shows that the use of randomized Q and R values did not degrade the performance of SKF and hence, one step closer to the realization of a parameter-less SKF.
Highlights
The simulated Kalman filter (SKF) [1] is a relatively new optimization algorithm compared to some well-known optimizer such as genetic algorithm (GA) [2], particle swarm optimization (PSO) [3], and gravitational search algorithm (GSA) [4]
The SKF has been hybridized with PSO and GSA [8,9]
In the proposed SKF with randomized Q and R values, Eq (2) is replaced with Eq (9) as follows: The experiment was conducted based on CEC2014 benchmark dataset for function minimization problems
Summary
The simulated Kalman filter (SKF) [1] is a relatively new optimization algorithm compared to some well-known optimizer such as genetic algorithm (GA) [2], particle swarm optimization (PSO) [3], and gravitational search algorithm (GSA) [4]. 2. The Original and Modified SKF Algorithms The maximum number of iterations, tmax, is defined. The initial value of error covariance estimate, (0) , the process noise value, , and the measurement noise value,
Sign-in/Register to view highlights & summary 
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call