Mirage search optimization: Application to path planning and engineering design problems

Abstract

In this article, a new meta-heuristic optimization algorithm motivated by mirage physical principles, named Mirage Search Optimization (MSO), is proposed. MSO mainly consists of two updating strategies, i.e., the superior mirage strategy and the inferior mirage strategy, which results in the global exploration and local exploitation capabilities, respectively. In addition, other two population evolution-guided mechanisms such as the fitness-distance balance (FDB) and fitness-distance constraint (FDC) are incorporated into MSO and termed as FDB-MSO and FDC-MSO, to further check and test the good optimization performance of MSO and its variants. MSO and 25 comparison algorithms are examined on CEC2017, CEC2014 and 21 classical benchmark functions. Optimization efficiency of MSO was verified by Wilcoxon rank sum test, Friedman test and stability analysis. Furthermore, competitiveness of MSO in solving real-world problems under constraints is demonstrated using six classical engineering problems. Finally, MSO is used for the path planning problem, which verifies applicability of MSO to real-world problems. Experimental results indicate MSO is competitive with other competing algorithms. Source codes of MSO are publicly available at https://www.mathworks.com/matlabcentral/fileexchange/180042-mirage-search-optimization.