On offset-free Model Predictive Control with Long Short-Term Memory Networks

This paper addresses offset-free reference tracking of asymptotically constant reference signals and disturbances using Model Predictive Control with a Long Short-Term Memory Network model. Basically, an error free output estimation and an error free nominal tracking for time-varying references are proposed for an Incremental Input-to-State Stable Long Short-Term Memory Network. This implies an offset-free Model Predictive Control algorithm.