Real-Time Optimal Control Using Particule Swarm Optimization
Resumen
Optimal Control Problems can be solved by a well known metaheuristic,
Particle Swarm Optimization. A version of this metaheuristic has been used to develop
an algorithm that solves such a problem. The solution can be used to control the
considered dynamic system only in open loop. Therefore, the main problem is how to
use the PSO algorithm in a control structure that works in closed loop. The solution of
this problem is to use the well known control structure Model Predictive Control whose
controller has an optimization algorithm that minimizes the prediction error. The
proposal of this work is to confer this role to the PSO based algorithm because it would
minimize implicitly the prediction error.