In this work we implemented a version of the Galactic Swarm Optimization metaheuristic algorithm tuned by a hidden Markov model. The Galactic Swarm Optimization algorithm is an abstraction of the motion of stars within galaxies on the first level, and galaxies within a cluster of galaxies on the second level. We address the problem of controlling the metaheuristic parameters by identifying the state of the algorithm at each iteration i, using the Hidden Markov Model framework and updating the Galactic Swarm Optimization parameters accordingly. The results obtained show an improvement compared to the original algorithm using the fixed parameters found in the literature. In addition, the results are compared against other algorithms that use different techniques and hybridizations to solve the same problem, showing an improvement in performance with a similar quality for the solutions obtained.