US 12,366,851 B2
System, device, and method for controlling a physical or chemical process
Petru Tighineanu, Ludwigsburg (DE); Attila Reiss, Renningen (DE); Felix Berkenkamp, Munich (DE); Julia Vinogradska, Gerlingen (DE); Kathrin Skubch, Frankfurt (DE); and Paul Sebastian Baireuther, Stuttgart (DE)
Assigned to ROBERT BOSCH GMBH, Stuttgart (DE)
Filed by Robert Bosch GmbH, Stuttgart (DE)
Filed on Sep. 22, 2022, as Appl. No. 17/950,542.
Claims priority of application No. 10 2021 210 774.5 (DE), filed on Sep. 27, 2021.
Prior Publication US 2023/0097371 A1, Mar. 30, 2023
Int. Cl. G05B 19/418 (2006.01); B25J 9/16 (2006.01); G05B 13/02 (2006.01); G05B 13/04 (2006.01); G05B 15/02 (2006.01)
CPC G05B 19/41885 (2013.01) [B25J 9/163 (2013.01); G05B 13/0265 (2013.01); G05B 13/042 (2013.01); G05B 15/02 (2013.01)] 8 Claims
OG exemplary drawing
 
1. A method for controlling a physical or chemical process, wherein each measurement point of known measurement points has an input parameter value of at least one input variable of the physical or chemical process and an output value associated with the input parameter value, of at least one output variable of the physical or chemical process, and wherein a first a posteriori model describes a relationship between at least one input variable and at least one output variable of another process related to the physical or chemical process, the method comprising:
determining a second a posteriori model using the first a posteriori model, wherein the second a posteriori model describes a relationship between the at least one input variable and at least one output variable of the physical or chemical process, wherein the determining of the second a posteriori model includes:
determining a plurality of Gaussian processes having a common covariance function, wherein each of the Gaussian processes is determined by drawing a function from the first a posteriori model and forming an expected value of the Gaussian process; and
(i) determining an a priori model as a mean value of the plurality of Gaussian processes and determining the second a posteriori model by optimizing hyperparameters of the model using the known measurement points in such a way that at least one hyperparameter of a covariance function of the first a posteriori model remains unchanged, and conditioning the model on the known measurement points, or (ii) conditioning each Gaussian process of the plurality of Gaussian processes on the known measurement points and determining the second a posteriori model as a mean value of the conditioned plurality of Gaussian processes; and
controlling the physical or chemical process using the second a posteriori model.