Computer-implemented method for configuring a virtual test system and training method

1 . A computer-implemented method for configuring a virtual test system for testing vehicle functions of a motor vehicle, the method comprising:

providing a dataset comprising first data of a plurality of input ports of an artifact under test and comprising second data of a plurality of output ports of at least one other artifact under test;

training a machine learning algorithm, the machine leaning algorithm comprising a neural network, using stochastic learning, to configure the virtual test system;

applying the machine learning algorithm to the dataset;

determining, for each of the plurality of input ports of the artifact under test, a confidence value for each of the plurality of output ports of the at least one other artifact under test;

assigning, for each of the plurality of input ports of the artifact under test, the output ports having the highest confidence value of the at least one other artifact under test depending on a first condition;

compiling a list of the output ports having the highest confidence values depending on a second condition or a non-assignment of an output port depending on a third condition in order to configure a connection of the input ports of the artifact under test to suitable output ports of the at least one other artifact under test;

based on the list of the output ports manually selecting, by a user, appropriate output ports;

connecting the appropriate output ports to appropriate input ports;

after said connecting the appropriate input ports of the artifact under test to the appropriate output ports of the at least one other artifact under test, performing a simulation of a software-based vehicle function; and

testing a vehicle function based on said simulation,

wherein the machine learning algorithm is trained to recognize which input port belongs to which output port, and

wherein the at least one other artifact under test is a virtual electronic control unit (ECU) integrated in a simulation system.

2 . The computer-implemented method according to claim 1 , wherein the first condition is met by the fact that if the confidence value is greater than or equal to a specified first threshold value or greater than or equal to 99%, and wherein the output port having the highest confidence value of the at least one other artifact under test is automatically assigned.

3 . The computer-implemented method according to claim 2 , wherein the second condition is met if the confidence value is less than the specified first threshold value or less than 99%, and greater than or equal to a specified second threshold value or greater than or equal to 80%, and wherein a list of a specified number of output ports having the highest confidence values is created.

4 . The computer-implemented method according to claim 3 , wherein a user manually selects the appropriate output port based on the compiled list of a specified number.

5 . The computer-implemented method according to claim 3 , wherein the confidence values included in the compiled list are sorted by the confidence values in ascending or descending order, and wherein the list includes confidence values which are less than the specified first threshold value or less than 99%, and greater than or equal to the specified second threshold value or greater than or equal to 80%.

6 . The computer-implemented method according to claim 3 , wherein the third condition is met if the confidence value is less than the specified second threshold value or less than 80%, and thereby no assignment of an output port is made.

7 . The computer-implemented method according to claim 1 , wherein the data set further comprises third data of a plurality of output ports of the artifact under test and fourth data of a plurality of input ports of at least of one other artifact under test.

8 . The computer-implemented method according to claim 7 , wherein the dataset contains data on all input ports and all output ports of the artifact under test and of the at least one other artifact under test or contains a user-made or automatically suggested selection of the input ports and output ports of the artifact under test and of the at least one other artifact under test.

9 . The computer-implemented method according to claim 1 , wherein each assignment of an output port of the at least one other artifact under test to an input port of the artifact under test is logged in a log file.

10 . The computer-implemented method according to claim 1 , wherein the artifact under test and the at least one other artifact under test are a virtual ECU and/or environmental model of the motor vehicle.

11 . The computer-implemented method according to claim 1 , wherein said training the machine learning algorithm comprises:

providing a first training dataset comprising first data of the plurality of input ports of the artifact under test and comprising second data of the plurality of output ports of the at least one other artifact under test;

providing a second training dataset comprising, for each of the plurality of input ports of the artifact under test, the confidence value for each of the plurality of output ports of the at least one other artifact under test; and

training the machine learning algorithm.

12 . The computer-implemented method according to claim 11 , wherein the optimization algorithm calculates an assignment of output ports of the at least one other artifact under test using names, alphabetic, alphanumeric designations, and/or metadata or a data type of the input ports of the artifact under test.

13 . The computer-implemented method according to claim 11 , wherein the confidence values of the second training dataset are less than a specified first threshold value or less than 99%, and greater than or equal to a specified second threshold value or greater than or equal to 80%.

14 . The computer-implemented method according to claim 11 , wherein the machine learning algorithm is trained based on an acceptance or non-acceptance by a user of an assignment performed by the machine learning algorithm of an output port having the highest confidence value or a non-assignment of an output port of the at least one other artifact under test.

15 . The computer-implemented method according to claim 1 , wherein said software-based vehicle function is an adaptive cruise control.