The toolbox provides tools for simulating your controller from the command line and in Simulink. If you are designing a controller using the MPC Designer app, you can simulate control scenarios during the design process and generate a Simulink model from your design.
|Compute optimal control action|
|Option set for mpcmove function|
|MPC controller state|
|Simulate closed-loop/open-loop response to arbitrary reference and disturbance signals for implicit or explicit MPC|
|MPC simulation options|
|Plot responses generated by MPC simulations|
|MPC Controller||Simulate model predictive controller|
|MPC Designer||Design and simulate model predictive controllers|
Simulate a model predictive controller with a nonlinear plant at the command line. At each control interval, relinearize the nonlinear plant and define a new controller based on the updated plant model.
Test an existing MPC controller within a Simulink model.
You can automatically generate a Simulink model that uses the current model predictive controller to control its internal plant model.
If your application allows you to anticipate trends in such signals, an MPC controller with signal previewing can improve reference tracking, measured disturbance rejection, or both.
Simulate an MPC controller when there is a mismatch between the controller prediction model and the actual plant dynamics.
You can update the constraints of your MPC controller at each control interval.
You can adjust the cost function penalty weights for your MPC controller while the controller operates.
You can adjust the prediction and control horizons for your MPC controller while the controller operates.
Reduce large actuator movements when changing controller operating modes.
You can switch between multiple MPC controllers based on their optimal objective function cost values.
You can detect controller failures in real time by using the optimization status controller output.
Simulate the closed-loop response of a model predictive controller with a custom quadratic programming solver.
You can guarantee the worst-case execution time for your MPC controller by applying a suboptimal solution after the number of optimization iterations exceeds a specified maximum value.
Design a model predictive controller in MATLAB and use cosimulation validate whether the controller is robust enough to control a nonlinear plant.