Regions of attraction and recursive feasibility in robust MPC

For linear systems with multiplicative uncertainty, a formulation of robust model predictive control in a lifted space is given and is proven to give better or equal performance to any tube MPC approach that uses the same parameterisation of the input. Then a tube MPC controller is formulated that ensures recursive feasibility through the use of terminal sets in the degrees of freedom available to the controller. This is shown to give performance comparable to the lifted approach while avoiding an exponential growth in computation.