Method for increasing control performance of model predictive control cost functions

Patent metadata

Field	Value
Publication number	US-11192561-B2
Application number	US-201916418658-A
Country	US
Kind code	B2
Filing date	May 21, 2019
Priority date	May 21, 2019
Publication date	Dec 7, 2021
Grant date	Dec 7, 2021

Abstract

A method for controlling an actuator system of a motor vehicle includes utilizing a model predictive control (MPC) module with an MPC solver to determine optimal positions of one or more actuators of the actuator system. The method further includes receiving a plurality of actuator system parameters, and triggering the MPC solver to generate one or more control commands from plurality of actuator system parameters. The method further includes applying a cost function to reduce a steady-state tracking error in the one or more control commands from the MPC solver and applying the one or more control commands to alter positions of the one or more actuators, and applying a penalty term to the steady-state predictions of positions of the plurality of actuators to limit a difference between a steady-state prediction of the actuator system and a solution from the MPC solver.

First claim

What is claimed is:

1. A method for controlling an actuator system of a motor vehicle, the method comprising:
utilizing a model predictive control (MPC) module with an MPC solver to determine optimal positions of one or more actuators of the actuator system;
receiving a plurality of actuator system parameters;
triggering the MPC solver to generate one or more control commands from the plurality of actuator system parameters;
computing a steady-state solution of the actuator system over a prediction horizon;
generating steady-state predictions of positions of the one or more actuators of the actuator system based on actuator system inputs, one or more environmental conditions, and a linearized physics-based model of the actuator system; and
applying a cost function to reduce a steady-state tracking error in the one or more control commands from the MPC solver; and
applying the one or more control commands to alter positions of the one or more actuators, and
applying a penalty term to the steady-state predictions of positions of the one or more actuators to limit a difference between a steady-state prediction of the actuator system and a solution from the MPC solver, wherein the penalty term has the following equation:

∑

i
=
1

n
⁢
y

⁢

W

y
,
i

⁡

(

y
i
s

-

y
i

r
⁢
e
⁢
f

)

2

where y s =CA −1 B+y n s , such that y s is a nominal steady-state solution to the actuator system, W y,i is a predetermined calibrated weight, B indicates the effects of one or more control inputs on a state of the actuator system, A expresses effects of a previous state on a current state, and C maps the current state of the actuator system to one or more output positions of the one or more actuators in the actuator system.

2. The method of claim 1 wherein triggering the MPC solver further comprises:
capturing a relationship between a steady-state response of the actuator system and one or more control inputs; and
calculating a steady-state gain for the actuator system.

3. The method of claim 2 wherein calculating a steady-state gain for the actuator system further comprises:
mapping states of the one or more actuators to output positions of the one or more actuators;
determining a relationship between a previous state of the one or more actuators and a current state of the one or more actuators; and
determining a relationship between the one or more control inputs and the current state of the one or more actuators.

4. The method of claim 1 wherein computing a steady-state solution of the actuator system further comprises:
determining a steady-state non-linearized solution of the actuator system based on a nominal input.

5. The method of claim 1 wherein receiving a plurality of actuator system parameters further comprises:
receiving one or more control inputs to the actuator system;
measuring one or more environmental conditions;
calculating a linearized physics-based model of the actuator system; and
determining one or more actuator system inputs to the linearized physics-based model.

6. The method of claim 5 further comprising:
generating one or more physical parameters from the linearized physics-based model.

7. The method of claim 1 wherein applying a cost function further comprises:
calculating the predetermined calibrated weight for each control command over the prediction horizon, wherein the prediction horizon is an infinite horizon.

8. The method of claim 7 wherein applying the one or more control commands to alter positions of the one or more actuators further comprises:
taking a measurement of the actuator system once the one or more control commands have been applied to the one or more actuators; and
applying the measurement of the actuator system to the linearized physics-based model.

9. The method of claim 8 wherein applying the measurement further comprises:
applying the measurement of the actuator system as a previous control input to the linearized physics-based model of the actuator system.

10. The method of claim 1 wherein controlling an actuator system further comprises:
controlling one or more of an air per cylinder (APC) system; a fuel system; a state of charge system; a heating, ventilation, and air conditioning system; and an advanced driver assistance system (ADAS).

11. A system for controlling an actuator system of a motor vehicle, the system comprising:
one or more of an air per cylinder (APC) system; a fuel system; a state of charge system; a heating, ventilation, and air conditioning system; and an advanced driver assistance system (ADAS);
one or more actuators;
a model predictive control (MPC) module with an MPC solver that determines optimal positions of the one or more actuators of the actuator system, the MPC module having a controller, a memory, and an input/output interface, the memory storing program code portions, and the controller configured to execute the program code portions, the program code portions comprising:
a program code portion that receives a plurality of actuator system parameters;
a program code portion that triggers the MPC solver to generate one or more control commands from plurality of actuator system parameters;
a program code portion that computes a steady-state solution of the actuator system over a prediction horizon;
a program code portion that generates steady-state predictions of positions of the one or more actuators of the actuator system based on the actuator system inputs, one or more environmental conditions, and a linearized physics-based model of the actuator system;
a program code portion that applies a cost function to reduce a steady-state tracking error in the one or more control commands from the MPC solver;
a program code portion that applies the one or more control commands via the input/output interface to alter positions of the one or more actuators; and
a program code portion that applies a penalty term to the steady-state predictions of positions of the one or more actuators to limit a difference between a steady-state prediction of the actuator system and a solution from the MPC solver, wherein the penalty term has the following equation:

∑

i
=
1

n
⁢
y

⁢

Assignees

• Gm Global Tech Operations Llc

Inventors

• Chen Jun
• Long Ruixing
• Hu Yiran

Classifications

• B60W50/0097 Primary

  Predicting future conditions · CPC title

• Y02T10/7072

  Electromobility specific charging systems or methods for batteries, ultracapacitors, supercapacitors or double-layer capacitors · CPC title

• B60L58/12

  responding to state of charge [SoC] · CPC title

• F02D41/26

  using computer, e.g. microprocessor · CPC title

• B60W10/08

  including control of electric propulsion units, e.g. motors or generators · CPC title