Method for testing a product

1. A computer-implemented method for testing a product, the product including software or hardware or a system including hardware and software, the method comprising the following steps:

in a first simulation categorization, for each of a plurality of tests of one or more simulations of the product:

carrying out, depending on input parameters, a respective one of the one or more simulations, using a particular property of the product being tested;

outputting, depending on a comparison between a result of the respective simulation and a requirement made of the particular property, a first classification as a result of the respective simulation;

determining, depending on a comparison between reference data from an alternative test of the particular property of the product and the requirement made of the particular property, a second classification; and

determining, based on a result of a comparison between the first classification and the second classification, a respective accuracy of the respective simulation;

executing, by a processor, machine learning using the accuracies determined in the first simulation categorization to train a simulation classifier; and

subsequent to the training of the simulation classifier, executing in a second simulation categorization, by the processor, the simulation classifier, by which execution the simulation classifier identifies an accuracy of a further simulation without comparing a classification made by the further simulation classifier to another classification made based on reference data.

2. The method as recited in claim 1 , wherein the reference data are from a test on the real product, or from a reference simulation.

3. The method as recited in claim 1 , wherein the first classification and the second classification each include whether the product passed or failed the test, or whether the test was not carried out because conditions specified for carrying out the test were not met.

4. The method as recited in claim 1 , wherein the particular property includes a diagnostic or detection function, and the first classification and the second classification each include whether the diagnostic or detection function gives a false positive, or a false negative, or a true positive, or a true negative result.

5. The method as recited in claim 1 , wherein the particular property comprises a safety-critical property of the product, and the first classification and the second classification each comprise whether a pre-defined, safety-relevant event occurred or not.

6. The method as recited in claim 1 , wherein the accuracy of the simulation is determined in the first simulation categorization using a confusion matrix or an information gain.

7. The method as recited in claim 1 , wherein the product includes a safety-critical software function or a safety-critical system for an at least partly automated vehicle or an at least partly automated robot.

8. A computer-readable storage medium on which is stored a computer program for testing a product, the product including software or hardware or a system including hardware and software, the computer-program, when executed by a computer, causing the computer to perform the following steps:

in a first simulation categorization, for each of a plurality of tests of one or more simulations of the product:

carrying out, depending on input parameters, a respective one of the one or more simulations, using a particular property of the product being tested;

outputting, depending on a comparison between a result of the respective simulation and a requirement made of the particular property, a first classification as a result of the respective simulation;

determining, depending on a comparison between reference data from an alternative test of the particular property of the product and the requirement made of the particular property, a second classification; and

determining, based on a result of a comparison between the first classification and the second classification, a respective accuracy of the respective simulation;

executing, by a processor, machine learning using the accuracies determined in the first simulation categorization to train a simulation classifier; and

subsequent to the training of the simulation classifier, executing in a second simulation categorization, by the processor, the simulation classifier, by which execution the simulation classifier identifies an accuracy of a further simulation without comparing a classification made by the further simulation classifier to another classification made based on reference data.

9. A test environment configured for testing a product, the product including software or hardware or a system including hardware and software, the test environment configured to:

in a first simulation categorization, for each of a plurality of tests of one or more simulations of the product:

carry out, depending on input parameters, a respective one of the one or more simulations, using a particular property of the product being tested;

output, depending on a comparison between a result of the respective simulation and a requirement made of the particular property, a first classification as a result of the respective simulation;

determine, depending on a comparison between reference data from an alternative test of the particular property of the product and the requirement made of the particular property, a second classification; and

determine, based on a result of a comparison between the first classification and the second classification, a respective accuracy of the respective simulation;

execute, by a processor of the environment, machine learning using the accuracies determined in the first simulation categorization to train a simulation classifier; and

subsequent to the training of the simulation classifier, execute in a second simulation categorization, by the processor, the simulation classifier, by which execution the simulation classifier identifies an accuracy of a further simulation without comparing a classification made by the further simulation classifier to another classification made based on reference data.