CONTROL DEVICE AND METHOD FOR CALCULATING AN OUTPUT PARAMETER FOR A CONTROLLER

• US 20130110749A1
• Filed: 04/06/2011
• Published: 05/02/2013
• Est. Priority Date: 04/27/2010
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Lang, Tobias, Markert, Heiner, Aue, Axel, Kruse, Thomas, Imhof, Volker, Richardsen, Thomas, Schulmeister, Ulrich, Bannow, Nico, Saetzler, Michael, Ulmer, Holger, Diener, Rene, Kloppenburg, Ernst, Streichert, Felix
• Granted Patent: US 10,013,658 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: F02D 2041/1433 using a model or simulation of the system
• CPC Class Codes: F02D 41/1401 characterised by the control or regulation method F02D41/1473, F02D41/1477 take precedence
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05D 1/0088 characterized by the autonomous decision making process, e.g. artificial intelligence, predefined behaviours using knowledge based models G06N5/00
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A control device in a vehicle includes a unit for calculating, during operation of the vehicle, on the basis of at least one input variable ascertained during operation, at least one output variable for a control system of functions of the vehicle. The control device performs the calculation of the output variables using a Bayesian regression of training values ascertained, before operation, for the output variable and the input variable.

Claims

1-12. -12. (canceled)

13. A control device in a vehicle, comprising:

a calculation unit configured to calculate, during operation of the vehicle and on the basis of at least one input variable ascertained during operation, at least one output variable for a control system of a function of the vehicle;

wherein the calculation unit is configured to perform the calculation of the output variable using a Bayesian regression over training values ascertained, before operation, for the output variable and the input variable.

14. The control device as recited in claim 13, wherein the calculation unit implements the Bayesian regression as one of a kriging, a Gaussian process, or a sparse Gaussian process.

15. The control device as recited in claim 14, wherein the calculation unit determines a variance, ascertained from the Bayesian regression, of the output variable.

16. The control device as recited in claim 15, further comprising:

means for initiating a corrective action if the ascertained variance exceeds a predetermined threshold value.

17. The control device as recited in claim 15, wherein for the calculation of the at least one output variable, a Bayesian regression dynamically modeled by autoregression is carried out over the training values ascertained before operation.

18. The control device as recited in claim 15, wherein the calculation unit includes a computation element which at least partly carries out the Bayesian regression.

19. The control device as recited in claim 18, wherein the calculation unit further includes a logic circuit, associated with the computation element, and wherein the logic circuit partly carries out the Bayesian regression.

20. The control device as recited in claim 19, wherein the logic circuit associated with the computation element is configured to calculate exponential functions.

21. A method for calculating by a control device in a vehicle, during operation of the vehicle and on the basis of at least one input variable ascertained during operation, at least one output variable for a control system of a function of the vehicle, comprising:

ascertaining training values for the output variable and the input variable before operation; and

performing the calculation of the output variable using a Bayesian regression over the training values.

22. The method as recited in claim 21, wherein a variance, ascertained from the Bayesian regression, of the output variable is also taken into consideration for the control system.

23. The method as recited in claim 22, wherein for the calculation of the at least one output variable, a Bayesian regression dynamically modeled by autoregression is carried out over the training values ascertained before operation.

24. The method as recited in claim 23, wherein the Bayesian regression is implemented as one of kriging, a Gauss process, or a sparse Gauss process.