CSGO: A Game-Inspired Metaheuristic Algorithm for Global Optimization

This paper presents a video game-inspired meta-heuristic algorithm and its performance evaluation. This optimizer algorithm is developed by assembling impressive features of previous well-known optimizer algorithms such as stochastic fractal search (SFS), artificial gorilla troops optimizer (GTO) and marine predators algorithm (MPA) with addition of chaotic operators. The main inspiration of this proposed chaotic SFS-GTO optimizer (CSGO) algorithm is the survival-of-the-fittest agent within a virtual map environment between two competitive groups in order to accomplish a mission using diverse strategies and information gathering-sharing activities. Then, the proposed CSGO's performance has been evaluated using thirteen standard benchmark test functions. The performance of CSGO is compared with its predecessors and latest improved grey wolf optimizer (MELGWO) algorithms. Based on the statistical and convergence curve analysis carried out, the proposed CSGO algorithm outperformed other competitor algorithms in terms of results accuracy and convergence speed with the exception of high computational time taken due to high number of function evaluations involved.