Moment Based Model Predictive Control for Systems with Additive Uncertainty

In this paper, we present a model predictive control (MPC) strategy based on the moments of the state variables and the cost functional. The statistical properties of the state predictions are calculated through the open loop iteration of dynamics and used in the formulation of MPC cost function. We show that the moment based formulation yields predictive control problems which are computationally simpler to solve compared to the existing robust MPC formulations, while providing statistical robustness properties. We apply the proposed MPC technique to a simple simulation example to demonstrate its effectiveness.