direct_data_driven_mpc.utilities.controller.controller_creation

direct_data_driven_mpc.utilities.controller.controller_creation#

Functions for creating data-driven MPC controllers.

This module provides functions to create data-driven MPC controllers for both LTI and nonlinear systems using controller configurations and initial input-output measurement data.

Functions

create_lti_data_driven_mpc_controller(...[, ...])

Create an LTIDataDrivenMPCController instance using a specified Data-Driven MPC controller configuration and initial input-output trajectory data measured from a system.

create_nonlinear_data_driven_mpc_controller(...)

Create a NonlinearDataDrivenMPCController instance using a specified Data-Driven MPC controller configuration and initial input-output trajectory data measured from a system.
create_lti_data_driven_mpc_controller(controller_config: LTIDataDrivenMPCParams, u_d: ndarray, y_d: ndarray, use_terminal_constraints: bool = True) → LTIDataDrivenMPCController[source]#

Create an LTIDataDrivenMPCController instance using a specified Data-Driven MPC controller configuration and initial input-output trajectory data measured from a system.

Parameters:

  • controller_config (LTIDataDrivenMPCParams) – A dictionary containing configuration parameters for a Data-Driven MPC controller designed for Linear Time-Invariant (LTI) systems.

  • u_d (numpy.ndarray) – An array of shape (N, m) representing a persistently exciting input sequence used to generate output data from the system. N is the trajectory length and m is the number of control inputs.

  • y_d (numpy.ndarray) – An array of shape (N, p) representing the system’s output response to u_d. N is the trajectory length and p is the number of system outputs.

  • use_terminal_constraints (bool) – If True, include terminal equality constraints in the Data-Driven MPC formulation. If False, the controller will not enforce these constraints. Defaults to True.

Returns:

LTIDataDrivenMPCController – An LTIDataDrivenMPCController instance, which represents a Data-Driven MPC controller designed for Linear Time-Invariant (LTI) systems, based on the specified configuration.

create_nonlinear_data_driven_mpc_controller(controller_config: NonlinearDataDrivenMPCParams, u: ndarray, y: ndarray) → NonlinearDataDrivenMPCController[source]#

Create a NonlinearDataDrivenMPCController instance using a specified Data-Driven MPC controller configuration and initial input-output trajectory data measured from a system.

Parameters:

  • controller_config (NonlinearDataDrivenMPCParams) – A dictionary containing configuration parameters for a Data-Driven MPC controller designed for nonlinear systems.

  • u (numpy.ndarray) – An array of shape (N, m) representing a persistently exciting input sequence used to generate output data from the system. N is the trajectory length and m is the number of control inputs.

  • y (numpy.ndarray) – An array of shape (N, p) representing the system’s output response to u. N is the trajectory length and p is the number of system outputs.

Returns:

NonlinearDataDrivenMPCController – A NonlinearDataDrivenMPCController instance, which represents a Data-Driven MPC controller designed for nonlinear systems, based on the specified configuration.

Contents