US 12,073,303 B2
Method for producing a test data record, method for testing, method for operating a system, apparatus, control system, computer program product, computer-readable medium, production and use
Jan Götz, Buckenhof (DE); and Andrés Botero Halblaub, Kirchseeon (DE)
Assigned to Siemens Industry Software NV, Leuven (BE)
Appl. No. 17/254,296
Filed by SIEMENS INDUSTRY SOFTWARE NV, Leuven (BE)
PCT Filed May 17, 2019, PCT No. PCT/EP2019/062774
§ 371(c)(1), (2) Date Dec. 20, 2020,
PCT Pub. No. WO2019/242955, PCT Pub. Date Dec. 26, 2019.
Claims priority of application No. 18178772 (EP), filed on Jun. 20, 2018.
Prior Publication US 2021/0216063 A1, Jul. 15, 2021
Int. Cl. G06N 3/08 (2023.01); G05B 19/418 (2006.01); G06F 18/214 (2023.01); G06N 3/044 (2023.01)
CPC G06N 3/044 (2023.01) [G05B 19/4183 (2013.01); G05B 19/41885 (2013.01); G06F 18/214 (2023.01); G06N 3/08 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method for generating at least one test data set for an algorithm that obtains image data as an input and outputs areas of action that are dependent upon the image data as an output, an algorithm that obtains sensor data as an input and outputs areas of action that are dependent upon the sensor data as an output, or a combination thereof, the method comprising:
providing a trained artificial intelligence;
stimulating the trained artificial intelligence; and
generating, by the stimulated artificial intelligence, at least one test data set that comprises image data and areas of action that are associated with the image data, sensor data and areas of action that are associated with the sensor data, or a combination thereof, an area of action of the areas of action that are associated with the image data or the areas of action that are associated with the sensor data being partially defined by a control parameter or a control value that lies within a desired region or span,
wherein providing the trained artificial intelligence comprises obtaining the trained artificial intelligence, obtaining the trained artificial intelligence comprising:
providing a learning data set that comprises real image data that is recorded by at least one image capturing facility and areas of action that are associated with the real image data, real sensor data that is recorded by at least one sensor facility and areas of action that are associated with the real sensor data, or a combination thereof;
providing an artificial intelligence; and
conditioning the artificial intelligence using the provided learning data set.