US 12,240,392 B2
Generating soft sensors for mechanical systems using sparse identification of nonlinear dynamics
Iron Tessaro, Paraná (BR); Leandro dos Santos Coelho, Paraná (BR); Viviana Cocco Mariani, Paraná (BR); and Roberto Zanetti Freire, Paraná (BR)
Assigned to VOLVO TRUCK CORPORATION, Gothenburg (SE)
Filed by Volvo Truck Corporation, Gothenburg (SE)
Filed on May 20, 2022, as Appl. No. 17/750,209.
Prior Publication US 2024/0001947 A1, Jan. 4, 2024
Int. Cl. B60R 16/023 (2006.01); B60W 50/00 (2006.01)
CPC B60R 16/023 (2013.01) [B60W 2050/0028 (2013.01); B60W 2510/06 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of generating a soft sensor for a mechanical system, the method comprising:
defining a training dataset of first sensor values received from at least one first sensor corresponding a first physical quantity;
training a Sparse Identification of Nonlinear Dynamics (SINDy) model using the training dataset to generate a mathematical function defining a relationship between the first sensor values and second soft sensor values corresponding to a second physical quantity; and
embedding machine readable instructions based on the mathematical function into an embedded system of the mechanical system comprising at least one first sensor, wherein the machine readable instructions, when executed by the embedded system, cause the embedded system to receive third sensor values corresponding to the first physical quantity and generate fourth soft sensor values based on the third sensor values and the mathematical function.