Computer-implemented method and system for determining odd-relevant scenarios in a data set

Abstract

The invention relates to a computer-implemented method for determining scenarios which are comprised by a data set of surroundings data of a motor vehicle and cover an operational design domain (10) of an automated driving function of the motor vehicle. The invention also relates to a computer-implemented method for carrying out a virtual test for validating an automated driving function of a motor vehicle. The invention further relates to a system (1) for determining scenarios which are comprised by a data set of surroundings data of a motor vehicle and cover an operational design domain (10) of an automated driving function of the motor vehicle.

Description

[0001] Description

[0002] title

[0003] Computer-implemented method and system for

[0004] Determining ODD-relevant scenarios in a data set

[0005] The present invention relates to a computer-implemented method for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle.

[0006] The present invention further relates to a computer-implemented method for performing a virtual test to validate an automated driving function of a motor vehicle.

[0007] Furthermore, the present invention relates to a system for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle.

[0008] State of the art

[0009] Driver assistance systems such as adaptive cruise control and / or functions for highly automated driving can be verified or validated using various testing methods. Simulations, in particular, can be used for this purpose. To create test scenarios for simulations, test drives must be conducted. The sensor data obtained in this way is then abstracted into a test scenario.

[0010] The input data is raw data, i.e., sensor data from real test drives in the form of recordings of radar echoes, 3D point clouds from lidar measurements, and image data. The output data are simulable driving scenarios that include both an environment and trajectories.

[0011] On the other hand, test scenarios can be generated synthetically, for example without prior test drives, or scenarios obtained from test drives can be augmented or modified to generate new test cases.

[0012] "Scenario Optimization for the Validation of Automated and Autonomous Driving Systems" (Florian Hauer, B. Holzmüller, 2019) discloses methods for the verification and validation of automated and autonomous driving systems, in particular the finding of suitable test scenarios for virtual validation.

[0013] The test methodology involves adapting a metaheuristic search to optimize scenarios. For this, a suitable search space and a suitable quality function must be established. Based on an abstract description of the system's functionality and use cases, parameterized scenarios are derived. However, from a certification perspective, to demonstrate sufficient validation of highly automated driving functions, test scenarios generated from sensor data from real test drives are preferable to synthetically generated test scenarios, as real test drives are often able to better represent the complexity of all variables influencing a test scenario.

[0014] The capabilities of the automated driving system are determined by defining its operational design domain (ODD). The operational design domain represents the operating environment in which an automated driving system can safely perform a dynamic driving task.

[0015] On the other hand, the synthetic generation of scenarios offers the advantage that it can generate test scenarios that a human domain expert might not consider, but which may be safety-critical.

[0016] Consequently, there is a need to improve existing methods for generating test scenarios that validate an automated driving function of a motor vehicle so that they provide a basis for reliable certification of coverage of the operational design domain of the driving function.

[0017] It is therefore an object of the invention to provide an improved

[0018] To provide a method for determining a test data-based coverage of the operational design domain of the automated driving function.

[0019] Disclosure of the invention

[0020] The object is achieved according to the invention by a computer-implemented method for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle, having the features of patent claim 1.

[0021] Furthermore, the object is achieved according to the invention by a computer-implemented method for carrying out a virtual test for validating an automated driving function of a motor vehicle with the features of patent claim 12.

[0022] The object is further achieved according to the invention by a system for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle, having the features of patent claim 13.

[0023] Furthermore, the object is achieved by a computer program product with a computer program that comprises software means for carrying out one of the methods according to the invention, wherein the computer program is executed on a computer, with the features of patent claim 14. Furthermore, the object is achieved by a computer-readable data carrier with program code of a computer program in order to carry out at least parts of one of the methods according to the invention when the computer program is executed on a computer, with the features of patent claim 15.

[0024] The invention relates to a computer-implemented method for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle.

[0025] The method comprises providing a first data set of concrete scenarios covering the operational design domain of the automated driving function of the motor vehicle, in particular synthetically generated ones.

[0026] Furthermore, the method comprises, for each concrete scenario of the first data set, determining positions and / or sections in geographical map data which meet predetermined requirements, in particular parameters, of the concrete scenario.

[0027] The method further comprises providing a second data set of sensor data of a journey of the motor vehicle detected by a plurality of vehicle-mounted environment recognition sensors and filtering the second data set using the positions and / or sections determined in the geographical map data.

[0028] Furthermore, the method comprises creating an object list of objects encompassed by the second data set at the filtered positions and / or sections of the geographical map data.

[0029] The method further comprises comparing first trajectory data for each concrete scenario encompassed by the first data set and second trajectory data from trajectory data encompassed by the object list, and, based on the comparison, determining scenarios encompassed by the second data set that cover the operational design domain of the automated driving function of the motor vehicle.

[0030] The invention further relates to a computer-implemented method for carrying out a virtual test for validating an automated driving function of a motor vehicle using the scenarios determined by the method according to the invention and covering the operational design domain of the automated driving function of the motor vehicle.

[0031] The invention further relates to a system for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle. The system comprises a first data provision unit configured to provide a first data set of specific scenarios, in particular synthetically generated ones, covering the operational design domain of the automated driving function of the motor vehicle.

[0032] In addition, the system comprises a first calculation unit which is configured to determine, for each concrete scenario of the first data set, positions and / or sections in geographical map data which meet predetermined requirements, in particular parameters, of the concrete scenario.

[0033] The system further comprises a second data provision unit which is configured to provide a second data set of sensor data of a journey of the motor vehicle detected by a plurality of vehicle-mounted environment recognition sensors, and a second calculation unit which is configured to filter the second data set using the positions and / or sections determined in the geographical map data.

[0034] Furthermore, the system comprises a third calculation unit which is configured to create an object list of objects encompassed by the second data set at the filtered positions and / or sections of the geographical map data, a fourth calculation unit which is configured to compare first trajectory data for each concrete scenario encompassed by the first data set and second trajectory data from trajectory data encompassed by the object list, and a fifth calculation unit which is configured to determine scenarios encompassed by the second data set covering the operational design domain of the automated driving function of the motor vehicle.

[0035] The invention further relates to a computer program product with a computer program comprising software means for carrying out one of the methods according to the invention, wherein the computer program is executed on a computer, and to a computer-readable data carrier with program code of a computer program in order to carry out at least parts of one of the methods according to the invention when the computer program is executed on a computer.

[0036] One idea of ​​the present invention is to enable coverage of the operational design domain of the automated driving function by the test scenarios comprised by the second data set by comparing trajectory data pairs of the first data set of concrete scenarios covering the operational design domain of the automated driving function of the motor vehicle, in particular synthetically generated ones, and the second data set of sensor data of the journey of the motor vehicle recorded by the plurality of vehicle-side environment recognition sensors.

[0037] This advantageously eliminates the need to implement scenario-specific detection logic to determine the coverage of the operational design domain of the automated driving function by the test scenarios encompassed by the second data set.

[0038] Likewise, map data are not required for the step of comparing the first data set and the second data set, since positions and / or sections of relevant test scenarios have already been determined in advance in geographical map data for the first data set.

[0039] Thus, it is possible to compare the first data set and the second data set to determine test scenarios of the second data set that cover the operational design domain of the automated driving function using only trajectory data pairs from the first and second data sets. This also contributes to an efficient determination of scenario coverage in the real data provided by the second data set.

[0040] The proposed method therefore severely limits the search for real data and thus ensures that trajectory data can be directly compared with one another without having to take underlying geographical information into account. This drastically simplifies scenario detection. Since less real data has to be processed by complex perception algorithms, the processing costs (typically cloud computing) are also reduced many times over. According to a preferred development of the invention, it is provided that a scenario covering the operational design domain of the automated driving function of the motor vehicle is determined if the first trajectory data comprised by the first data set have at least a predetermined degree of similarity to the second trajectory data comprised by the object list.

[0041] As a result, the scenarios covered by the first data set and the second data set do not have to completely overlap. Applying the criterion of requiring a predefined similarity measure thus allows a certain degree of flexibility in determining relevant test scenarios covered by the second data set that fulfill the operational design domain of the motor vehicle's automated driving function.

[0042] According to a further preferred development of the invention, it is provided that the first trajectory data are transformed into a first feature vector and the second trajectory data are transformed into a second feature vector, wherein a first machine learning algorithm is applied to a trajectory data pair comprising the first feature vector of a respective concrete scenario encompassed by the first data set and the second feature vector of trajectory data encompassed by the object list. By representing the trajectory data in a feature space, an objective comparison basis for these can be created.

[0043] According to a further preferred development of the invention, it is provided that the first machine learning algorithm determines, in particular classifies, based on a vector distance between the first feature vector and the second feature vector, whether the second tra ctory data comprised by the second data set cover the operational design domain of the automated driving function of the motor vehicle.

[0044] A classification has the advantage of a definitive statement or determination as to whether the second tra ctory data comprised by the second data set cover the operational design domain of the automated driving function of the motor vehicle.

[0045] According to a further preferred development of the invention, it is provided that the filtering of the second data set using the positions and / or sections determined in the geographical map data comprises determining geographical positions and / or sections encompassed by the second data set which meet the predetermined requirements, in particular the predetermined parameters, of the respective concrete scenario.

[0046] Thus, it can be advantageously determined whether the positions and / or sections determined in the geographical map data are also present in the second data set, which additionally contain the specified

[0047] Fulfill the parameters of the respective concrete scenario.

[0048] According to a further preferred development of the invention, it is provided that the predetermined requirements, in particular the predetermined parameters, of the concrete scenario comprise a number of lanes, a lane width and / or a road curvature.

[0049] The specified parameters of the respective concrete scenario thus enable comparison with the real data of the second data set in order to identify whether the concrete scenarios of the first data set are at least partially covered by the second data set.

[0050] According to a further preferred development of the invention, it is provided that the filtering of the second data set is carried out using the positions and / or sections determined in the geographical map data by means of a predetermined similarity metric.

[0051] As a result, the geographical map data and the scenarios comprised by the second data set do not have to completely coincide. The application of the predetermined similarity metric thus enables flexibility in determining relevant test scenarios comprised by the second data set which fulfill the operational design domain of the automated driving function of the motor vehicle. According to a further preferred development of the invention, the second data set is searched using the positions and / or sections determined in the geographical map data for sequences at which the motor vehicle, in particular an ego vehicle, drove on road sections fulfilling the predetermined similarity metric.

[0052] This advantageously allows sequences from the second data set to be used which correspond to a certain degree with the positions and / or sections determined in the geographical map data.

[0053] According to a further preferred development of the invention, it is provided that the object list is created for each sequence that satisfies the predetermined similarity metric.

[0054] The object lists are an automatically recognized environment of the motor vehicle, which is stored in a structured manner for each magazine.

[0055] Based on the object list, what happened around the ego vehicle can then be reconstructed. This can be translated into a three-dimensional environment. The detected objects can be projected onto road network data.

[0056] According to a further preferred development of the invention, the creation of the object list is carried out using a second machine learning algorithm, which is configured to determine the objects encompassed by the second data set at the filtered positions and / or sections of the geographical map data through object detection. Thus, the object lists can be created automatically based on the second data set.

[0057] According to a further preferred development of the invention, object list data of the respective object list are located in a coordinate system of the motor vehicle having the plurality of vehicle-mounted environment detection sensors. Thus, a trajectory of the ego vehicle can be generated or reconstructed in the coordinate system with or without map or road network data.

[0058] The features of the computer-implemented method described herein for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle are likewise applicable to the system according to the invention for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle, and vice versa.

[0059] Brief description of the drawings For a better understanding of the present invention and its advantages, reference is now made to the following description in conjunction with the accompanying drawings.

[0060] The invention is explained in more detail below with reference to exemplary embodiments which are shown in the schematic illustrations of the drawings.

[0061] It shows :

[0062] Fig. 1 is a flow diagram of a computer-implemented method for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle according to a preferred embodiment of the invention; and

[0063] Fig. 2 is a schematic representation of a system for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain of an automated driving function of the motor vehicle according to the preferred embodiment of the invention.

[0064] Unless otherwise indicated, like reference symbols refer to like elements in the drawings.

[0065] Detailed description of the embodiments The computer-implemented method shown in Fig. 1 for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain 10 of an automated driving function of the motor vehicle comprises providing S 1 a first data set 12 of concrete scenarios 12a covering the operational design domain 10 of the automated driving function of the motor vehicle, in particular synthetically generated.

[0066] The method further comprises, for each specific scenario 12a of the first data set 12, determining S2 positions 14a and / or sections 14b in geographical map data 14 which satisfy predetermined requirements, in particular parameters, of the specific scenario 12a, and providing S3 a second data set 16 of sensor data of a journey of the motor vehicle recorded by a plurality of vehicle-mounted environment recognition sensors.

[0067] Furthermore, the method comprises filtering S4 the second data set 16 using the positions 14a and / or sections 14b determined in the geographical map data 14 and creating S5 an object list 18 of objects encompassed by the second data set 16 at the filtered positions 14a and / or sections 14b of the geographical map data 14.

[0068] The method also includes comparing S 6 first

[0069] Traj ectory data 20a for each through the first

[0070] Data set 12a comprised concrete scenario 12a and second trajectory data 20b of trajectory data comprised by the object list 18 and a determination S7 of scenarios comprised by the second data set 16 covering the operational design domain 10 of the automated driving function of the motor vehicle.

[0071] Furthermore, a scenario covering the operational design domain 10 of the automated driving function of the motor vehicle is determined if the first trajectory data 20a comprised by the first data set 12a have at least a predetermined degree of similarity to the second trajectory data 20b comprised by the object list 18.

[0072] The first trajectory data 20a are transformed into a first feature vector 19a and the second trajectory data 20b are transformed into a second feature vector 19b. A first machine learning algorithm A1 is thereby applied to a

[0073] Traj ectory data pair comprising the first feature vector 19a of a respective concrete scenario 12a comprised by the first data set 12a and the second feature vector 19b of traj ectory data comprised by the object list 18 is applied.

[0074] The first machine learning algorithm A1 determines, in particular further classifies based on a vector distance between the first feature vector 19a and the second feature vector 19b, whether the second trajectory data 20b comprised by the second data set 16 covers the operational design domain 10 of the automated

[0075] Cover the driving function of the motor vehicle. The filtering S4 of the second data set 16 using the positions 14a and / or sections 14b determined in the geographical map data 14 comprises determining geographical positions 14a and / or sections 14b encompassed by the second data set 16 which meet the predetermined requirements, in particular the predetermined parameters, of the respective concrete scenario 12a.

[0076] The specified requirements, in particular the specified parameters, of the concrete scenario 12a include a number of lanes, a lane width and / or a road curvature.

[0077] The filtering S4 of the second data set 16 is carried out using the positions 14a and / or sections 14b determined in the geographical map data 14 by means of a predetermined similarity metric 24.

[0078] The second data set 16 is further searched using the positions 14a and / or sections 14b determined in the geographical map data 14 for sequences 22 at which the motor vehicle, in particular an ego vehicle, drove on road sections 14b satisfying the predetermined similarity metric 24, wherein the object list 18 is created for each sequence 22 satisfying the predetermined similarity metric 24.

[0079] The creation of the object list 18 is carried out using a second machine learning algorithm A2, which is configured to determine the objects encompassed by the second data set 16 at the filtered positions 14a and / or sections 14b of the geographical map data 14 by object detection.

[0080] Object list data of the respective object list 18 are located in a coordinate system 1 of the motor vehicle having the majority of vehicle-side environment detection sensors.

[0081] Furthermore, a virtual test is carried out to validate the automated driving function of the motor vehicle using the scenarios determined by a method according to the invention and covering the operational design domain 10 of the automated driving function of the motor vehicle.

[0082] Fig. 2 shows a schematic representation of a system 1 for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain 10 of an automated driving function of the motor vehicle according to the preferred embodiment of the invention.

[0083] The system 1 comprises a first data provision unit 26, which is configured to provide a first data set 12a of concrete scenarios 12a covering the operational design domain 10 of the automated driving function of the motor vehicle, in particular synthetically generated ones. In addition, the system 1 comprises a first calculation unit 28, which is configured to determine, for each concrete scenario 12a of the first data set 12, positions 14a and / or sections in geographical map data 14 which satisfy predetermined requirements, in particular parameters, of the concrete scenario 12a.

[0084] The system 1 further comprises a second data provision unit 30 which is configured to provide a second data set 16 of sensor data of a journey of the motor vehicle detected by a plurality of vehicle-mounted environment recognition sensors, and a second calculation unit 32 which is configured to filter the second data set 16 using the positions 14a and / or sections 14b determined in the geographical map data 14.

[0085] Furthermore, the system 1 comprises a third calculation unit 34 which is configured to create an object list 18 of objects encompassed by the second data set 16 at the filtered positions 14a and / or sections 14b of the geographical map data 14, a fourth calculation unit 36 ​​which is configured to compare first trajectory data for each concrete scenario 12a encompassed by the first data set 12a and second trajectory data from trajectory data encompassed by the object list 18, and a fifth calculation unit 38 which is configured to determine scenarios encompassed by the second data set 16 covering the operational design domain 10 of the automated driving function of the motor vehicle.

[0086] Although specific embodiments have been illustrated and described herein, it will be understood by those skilled in the art that numerous alternative and / or equivalent implementations exist. It should be noted that the exemplary embodiment or exemplary embodiments are merely examples and are not intended to limit the scope, applicability, or configuration in any way.

[0087] Rather, the foregoing summary and detailed description will provide one skilled in the art with a convenient road map for implementing at least one exemplary embodiment, it being understood that various changes in the functionality and arrangement of elements may be made without departing from the scope of the appended claims and their legal equivalents.

[0088] In general, this application is intended to cover modifications, adaptations, or variations of the embodiments presented herein.

[0089] For example, the sequence of the method steps can be changed. Furthermore, the methods according to the invention can be carried out sequentially or in parallel, at least in sections. List of reference symbols

[0090] 1 system

[0091] 10 Operational Design Domain

[0092] 12 first data set

[0093] 12a concrete S scenario

[0094] 14 geographical map data

[0095] 14a positions

[0096] 14b Sections

[0097] 16 second data set

[0098] 18 Object list

[0099] 19a first feature vector

[0100] 19b second feature vector

[0101] 20a first tra ctory data

[0102] 20b second tra ctory data

[0103] 22 sequences

[0104] 24 Similarity metric

[0105] 26 first data provision unit

[0106] 28 first calculation unit

[0107] 30 second data provision unit

[0108] 32 second calculation unit

[0109] 34 third calculation unit

[0110] 36 fourth calculation unit

[0111] 38 fifth calculation unit

[0112] The first algorithm of machine

[0113] Learning

[0114] A2 second algorithm of the machine

[0115] Learning

[0116] S 1-S7 process steps

Claims

Claims 1. A computer-implemented method for determining scenarios comprising a data set of environmental data of a motor vehicle and covering an operational design domain (10) of an automated driving function of the motor vehicle, comprising the steps of: providing (S1) a first data set (12) of concrete scenarios (12a) covering the operational design domain (10) of the automated driving function of the motor vehicle, in particular synthetically generated; for each concrete scenario (12a) of the first data set (12), determining (S2) positions (14a) and / or sections (14b) in geographical map data (14) which meet predetermined requirements, in particular parameters, of the concrete scenario (12a); Providing (S3) a second data set (16) of sensor data of a journey of the motor vehicle detected by a plurality of vehicle-mounted environment detection sensors; Filtering (S4) the second data set (16) using the positions (14a) and / or sections (14b) determined in the geographical map data (14); Creating (S5) an object list (18) of objects encompassed by the second data set (16) at the filtered positions (14a) and / or sections (14b) of the geographical map data (14); Comparing (S6) first trajectory data (20a) for each concrete scenario (12a) encompassed by the first data set (12a) and second trajectory data (20b) of trajectory data encompassed by the object list (18); and based on the comparison, determining (S7) scenarios encompassed by the second data set (16) that cover the operational design domain (10) of the automated driving function of the motor vehicle.

2. Computer-implemented method according to claim 1, wherein a scenario covering the operational design domain (10) of the automated driving function of the motor vehicle is determined if the first Traj ectory data (20a) have at least one predetermined degree of similarity to the second traj ectory data (20b) comprised by the object list (18).

3. Computer-implemented method according to claim 2, wherein the first trajectory data (20a) are transformed into a first feature vector (19a) and the second trajectory data (20b) are transformed into a second feature vector (19b), wherein a first machine learning algorithm (A1) is applied to a trajectory data pair comprising the first feature vector (19a) of a respective concrete scenario (12a) comprised by the first data set (12a) and the second feature vector (19b) of trajectory data comprised by the object list (18).

4. The computer-implemented method according to claim 3, wherein the first machine learning algorithm (Al) determines, in particular classifies, based on a vector distance between the first feature vector (19a) and the second feature vector (19b), whether the second trajectory data (20b) comprised by the second data set (16) cover the operational design domain (10) of the automated driving function of the motor vehicle.

5. Computer-implemented method according to one of the preceding claims, wherein the filtering (S4) of the second data set (16) using the positions (14a) and / or sections (14b) determined in the geographical map data (14) comprises determining geographical positions (14a) and / or sections (14b) encompassed by the second data set (16) which meet the predetermined requirements, in particular the predetermined parameters, of the respective concrete scenario (12a).

6. Computer-implemented method according to one of the preceding claims, wherein the predetermined requirements, in particular the predetermined parameters, of the concrete scenario (12a) comprise a number of lanes, a lane width and / or a road curvature.

7. Computer-implemented method according to one of the preceding claims, wherein the filtering (S4) of the second data set (16) using the positions (14a) and / or sections (14b) determined in the geographical map data (14) by means of a given similarity metric (24) is carried out.

8. Computer-implemented method according to claim 7, wherein the second data set (16) is searched using the positions (14a) and / or sections (14b) determined in the geographical map data (14) for sequences (22) at which the motor vehicle, in particular an ego vehicle, drove on road sections (14b) satisfying the predetermined similarity metric (24).

9. Computer-implemented method according to claim 8, wherein the object list (18) is created for each sequence (22) satisfying the predetermined similarity metric (24).

10. The computer-implemented method according to claim 9, wherein the creation of the object list (18) is performed using a second machine learning algorithm (A2) which is configured to determine the objects encompassed by the second data set (16) at the filtered positions (14a) and / or sections (14b) of the geographical map data (14) by object detection.

11. Computer-implemented method according to one of the preceding claims, wherein object list data of the respective object list (18) are located in a coordinate system (1) of the motor vehicle having the plurality of vehicle-side environment recognition sensors.

12. Computer-implemented method for performing a virtual test to validate an automated driving function of a motor vehicle using the scenarios determined by a method according to one of claims 1 to 11 and covering the operational design domain (10) of the automated driving function of the motor vehicle.

13. System (1) for determining scenarios encompassed by a data set of environmental data of a motor vehicle and covering an operational design domain (10) of an automated driving function of the motor vehicle, comprising: a first data provision unit (26) configured to provide a first data set (12a) of concrete scenarios (12a) covering the operational design domain (10) of the automated driving function of the motor vehicle, in particular synthetically generated; a first calculation unit (28) configured to determine, for each concrete scenario (12a) of the first data set (12), positions (14a) and / or sections in geographical map data (14) which satisfy predetermined requirements, in particular parameters, of the concrete scenario (12a);a second data provision unit (30) which is configured to provide a second data set (16) of sensor data of a journey of the motor vehicle detected by a plurality of vehicle-side environment detection sensors; a second calculation unit (32) configured to filter the second data set (16) using the positions (14a) and / or sections (14b) determined in the geographical map data (14); a third calculation unit (34) configured to create an object list (18) of objects encompassed by the second data set (16) at the filtered positions (14a) and / or sections (14b) of the geographical map data (14); a fourth calculation unit (36) configured to compare first trajectory data for each specific scenario (12a) encompassed by the first data set (12a) and second trajectory data of trajectory data encompassed by the object list (18); and a fifth calculation unit (38) which is configured to determine, by means of the second data set (16), scenarios covering the operational design domain (10) of the automated driving function of the motor vehicle.

14. A computer program product comprising a computer program comprising software means for carrying out one of the methods according to any one of claims 1 to 11 and 12, wherein the computer program is executed on a computer.

15. Computer-readable data carrier with program code of a computer program to carry out at least parts of one of the methods according to one of claims 1 to 11 and 12 to run if the computer program is on a computer is running.

Images