Method for carrying out data processing

1 . A method for carrying out data processing, the method comprising:

creating a first computer program for a predetermined data processing task by one or more human programmers, wherein the predetermined data processing task ascertains a processing result of a control signal for a robotic device;

creating a plurality of second computer programs for the predetermined data processing task, wherein each of the computer programs is created by an artificial intelligence;

concurrently performing:

processing an input by the first computer program to ascertain a first processing result; and

processing the input by each of the plurality of second computer programs to ascertain a respective second processing result;

ascertaining a number of computer programs among the second computer programs whose second processing result contradicts the first processing result;

checking whether the number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than a predetermined threshold, which is greater than or equal to two;

initiating a safety measure when the number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than the predetermined threshold; and

in response to the number of computer programs among the second computer programs whose second processing result contradicts the first processing result being greater than the predetermined threshold, performing, by the data processing system, a safety measure comprising: discarding the first processing result from a memory and using a predetermined safe default value as an end result control signal of the data processing, wherein the default value controls the robotic device into a safe state preventing any damage or hazard from occurring.

2 . The method according to claim 1 , further comprising using the first processing result as a result of the processing when the ascertained number is less than the predetermined threshold.

3 . The method according to claim 1 , further comprising:

for each input of several inputs:

processing the input by the first computer program to ascertain a first processing result for the input, and processing the input by each of the plurality of second computer programs to ascertain a respective second processing result for the input; and

excluding one of the second computer programs from processing subsequent inputs, wherein the excluded second computer program has produced, for a number of inputs that exceeds a further predetermined threshold, a second processing result that contradicts the respective first processing result, wherein the number of computer programs among the second computer programs whose second processing result for the input contradicts the first processing result is not greater than the predetermined threshold.

4 . The method according to claim 3 , further comprising:

re-training the artificial intelligence when one of the second computer programs has in each case produced, for a number of inputs of several inputs that exceeds the further predetermined threshold, a second processing result that contradicts the respective first processing result, although the number of computer programs among the second computer programs whose second processing result for the input contradicts the first processing result is not greater than the predetermined threshold.

5 . The method according to claim 1 , further comprising:

for each input of several inputs,

selecting a portion of the plurality of second computer programs;

processing the input by the first computer program to ascertain a first processing result, and

processing the input by each second computer programs of the selected portion to ascertain a respective second processing result;

ascertaining a number of computer programs among the second computer programs of the selected portion whose second processing result contradicts the first processing result;

checking whether the number of computer programs among the second computer programs of the selected portion whose second processing result contradicts the first processing result is greater than a predetermined threshold, which is greater than or equal to two; and

initiating a safety measure if the number of computer programs among the second computer programs of the selected portion whose second processing result contradicts the first processing result is greater than the predetermined threshold.

6 . The method according to claim 1 , wherein the safety measure includes verifying a correct functioning of the first computer program.

7 . A data processing system having a processor configured to:

create a first computer program for a predetermined data processing task by one or more human programmers, wherein the predetermined data processing task ascertains a processing result of a control signal for a robotic device;

create a plurality of second computer programs for the predetermined data processing task wherein each of the computer programs is created by an artificial intelligence;

concurrently perform:

process an input by the first computer program to ascertain a first processing result, and process the input by each of the plurality of second computer programs to ascertain a respective second processing result;

ascertain a number of computer programs among the second computer programs whose second processing result contradicts the first processing result;

check whether a number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than a predetermined threshold, which is greater than or equal to two;

initiate a safety measure if the number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than the predetermined threshold; and

in response to the number of computer programs among the second computer programs whose second processing result contradicts the first processing result being greater than the predetermined threshold, perform, by the data processing system, a safety measure comprising: discarding the first processing result from a memory and using a predetermined safe default value as an end result control signal of the data processing, wherein the default value controls the robotic device into a safe state preventing any damage or hazard from occurring.

8 . A non-transitory computer-readable medium on which are stores instructions that, when executed by a processor, cause the processor to perform the following steps:

creating a first computer program for a predetermined data processing task by one or more human programmers, wherein the predetermined data processing task ascertains a processing result of a control signal for a robotic device;

creating a plurality of second computer programs for the predetermined data processing task, wherein each of the computer programs is created by an artificial intelligence;

concurrently performing:

processing an input by the first computer program to ascertain a first processing result, and processing the input by each of the plurality of second computer programs to ascertain a respective second processing result;

ascertaining a number of computer programs among the second computer programs whose second processing result contradicts the first processing result;

checking whether the number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than a predetermined threshold, which is greater than or equal to two;

initiating a safety measure if the number of computer programs among the second computer programs whose second processing result contradicts the first processing result is greater than the predetermined threshold; and

in response to the number of computer programs among the second computer programs whose second processing result contradicts the first processing result being greater than the predetermined threshold, performing, by the data processing system, a safety measure comprising: discarding the first processing result from a memory and using a predetermined safe default value as an end result control signal of the data processing, wherein the default value controls the robotic device into a safe state preventing any damage or hazard from occurring.