Remarks on an Echo State Network–Based Optimal Predictive Control Using a Metaheuristics Optimisation Approach

In this study, we investigated the capability of applying reservoir computing, especially echo state networks (ESNs), to control dynamic systems. To achieve optimal predictive control (OPC) of nonlinear systems, an ESN was used as the prediction model of the objective plant in a model predictive control approach. We performed computational experiments of the ESN–based OPC to test the tracking tasks of the discrete–time nonlinear plant. The simulation results showed that the ESN–based OPC could perform the control task with applicable accuracy, demonstrating the potential application of ESNs for controlling nonlinear systems.