Is it possible to implement nonlinear arx in nonlinear mpc

I understand that MATLAB allows the use of nonlinear prediction model in mpc by specifying the state and output functions. For nonlinear ARX, its states are "an ordered list of delayed input and output variables that define the structure of the model regressors" (https://www.mathworks.com/help/ident/ref/idnlarx.html). So, if my nonlinear ARX has three regressors (y(t-1), y(t-2) and u(t-1)), is what I'm doing here correct where I set the state function as equals to x(1:3) and output function as equals to x(4):

Answer 1

[Suresh Maddina]

Hi, it is my understanding that you are trying to use nonlinear ARX model for prediction model within Nonlinear MPC Simulink in Matlab.

The prediction model inside the nonlinear MPC must be a linear or nonlinear state & output functions of the form specified here: https://www.mathworks.com/help/mpc/ug/specify-prediction-model-for-nonlinear-mpc.html

[aurochs]

Hi Suresh,

Thanks for the answer.

I've already gone through the link before but I couldn't find any clue how to implement the system in my question.

I would appreciate it if there's any specific example on how to implement nonlinear arx model in the nonlinear mpc controller.

Thanks.

[Suresh Maddina]

In case you are referring to the prediction model inside the Nonlinear MPC, it does not support the ARX models. Only the state-space model of the form is supported mentioned in above documentation link.

For the plant model there are no restrictions on the type of model, so you can use the ARX model

[aurochs]

I'm referring to the prediction model. Thanks for the answer.