Specifying reference state values in MPC controller cost function

Hello,

I have an MPC controller, built with the following general format (i have not included the nonlinear equations because they are relatively long):

nx = 11;
ny = 3;
nu = 13;
nlobj = nlmpc(nx,ny,nu);

nlobj.Model.StateFcn="nonlinear_eom.m";
nlobj.Jacobian.StateFcn="nonlinear_j";
Ts=0.4;
p=30;
nlobj.Ts=Ts;
nlobj.PredictionHorizon=p;
nlobj.ControlHorizon=p;

I want to use this controller as part of a simulation where a preliminary value of the control values U(5:13) for each time step is to be given to the controller i.e. I want to specify a cost function where given the reference controls (using subscripts U), the cost is given by:

I looked at the example in Trajectory Optimization and Control of Flying Robot Using Nonlinear MPC - MATLAB & Simulink (mathworks.com)

This only considers the control values themselves. I know that:

nlmpcmove(nlmpcobj,x,lastmv,ref)

includes ref, which is the output reference values, but what about the input reference values?