Method for comparing plurality of data sets

Abstract

The present invention relates to a method for comparing a plurality of data sets, where each of the plurality of data sets respectively comprises one or more attributes, the method comprising: providing for each of the plurality of data sets respectively a comparison data set characterizing the respective data set, where each comparison data set respectively has one or more comparison attributes; assigning coordinates of the coordinate representation to one or each of a plurality of comparison attributes for each of the plurality of comparison data sets using a predetermined common coordinate representation, and determining a respective coordinate value of the respective coordinate assigned to the comparison attribute on the basis of a value of the respective comparison attribute; determining a common convex envelope of all coordinate values of the plurality of comparison data sets; determining the volume of the convex envelope; determining, on the basis of a comparison of the volume of the convex envelope with the comparison volume, whether the plurality of data sets are assumed to be identical or assumed to be identical within a predetermined tolerance as a comparison result; and providing a comparison result for further use.

Description

Technical Field

[0001] The invention relates to a method for comparing a plurality of data sets and to a computing system and a computer program for performing the method. Background Art

[0002] In various fields, for example for vehicles, data such as sensor data are processed, ie various calculations are performed; the results can then be used again in the vehicle, for example. Summary of the invention

[0003] According to the invention, a method for comparing a plurality of data sets and a computing system and a computer program for carrying out the method are proposed having the features of the independent patent claims. Advantageous embodiments are the subject matter of the dependent claims and the following description.

[0004] The present invention generally deals with the processing of data and in particular with the comparison of a plurality of data sets. In various fields, for example in the case of vehicles or generally so-called infrastructure-based autonomous driving, data, such as sensor data of sensors of vehicles or traffic infrastructure objects, are acquired and processed, i.e., various calculations are performed, for example; the results can then be used again, for example in the vehicle or in other traffic infrastructure objects.

[0005] In the context of partially automated driving or automated driving, complex calculations using sensor data are necessary, for example, in order to be able to assess what reaction should be carried out in the vehicle. As mentioned, due to the high volume of data and often complex calculations, it is expedient in this context to offload such processing to a vehicle-external, possibly distributed, efficient computing system. For this purpose, the data can be transmitted from the vehicle to a corresponding computing system (e.g., a computer center) via a suitable, usually wireless, communication connection. There, the data are then processed accordingly and, if necessary, subsequently transmitted back to the vehicle.

[0006] For many applications, especially even in vehicles, it is appropriate or often necessary to ensure a certain safety regarding the correctness of the processed data or calculation processes. In other words, possible errors in processing or calculation should be excluded as much as possible and the probability of this should be minimized.

[0007] In order to ensure a certain safety when calculating the data, the results or result data sets can be calculated from the original data or original data sets in different paths, respectively, wherein the most identical results should be obtained in all paths.

[0008] A principle that can be used for this is the so-called lockstep method. In the field of computer technology, in particular in the case of processors and microcontrollers, this refers to a method for fault tolerance and error detection in hardware, which is achieved by using a plurality of identical or similar units, such as CPU cores in a multi-core processor. For example, processors operating in lockstep are used in safety-critical applications, such as engine control units (ECUs) in vehicles and autonomous drive control (FADEC) in aircraft.

[0009] This approach can also be implemented on the basis of software, for example in a computing system. For example, such a software-based lockstep solution can be used according to the safety standards ISO 26262 and IEC 61508 for occasional errors in arithmetic operations. In this case, the data to be processed are processed in a plurality of, for example, two, usually even three or more separate paths according to a predetermined processing scheme (the desired calculation is performed, that is, the same calculation is performed in both paths). The data processed in the two paths are then subjected to a plausibility check or comparison with each other within the scope of a plausibility check. Errors can be detected using this approach.

[0010] For example, DE 10 2021 202 935 A1 describes a basic functional approach for processing data, for example in the cloud, within the scope of a lockstep method.

[0011] Usually, therefore in each processing path in a plurality of processing paths, respectively determine the result data set or generally speaking the data set from the original data set.Here, the processing paths are different from each other.As in the execution type (such as programming language, algorithm, tool configuration), these processing paths can be different, for example, in hardware (such as different processors) and / or software equipment (such as operating system, platform) in exactly the same way.Thereby derive a plurality of result data sets or generally speaking the data sets that must be compared with each other.If the result data sets are all consistent (that is, the same) or are similar in a predetermined range, that is, the same within a predetermined tolerance, the result can be considered to be correct, and there is no doubt that it is distorted due to systematic or occasional errors.Usually allow a certain tolerance, because slightly different results may occur due to different methods, libraries, algorithms, etc.

[0012] However, similar results are usually very expensive. For performance reasons, it is desirable to keep the number of comparisons low, but this requires a mathematical equality relationship. In a mathematical sense, such a relationship must be transferable. But this is not given, for example, in the case of position comparisons and, for example, acceptable spacing margins. For example, object A can have an acceptable distance d with a second object B, and B can, for example, have an acceptable distance with a third object C, and nevertheless A and C are too far apart from each other (for example, spacing 2d). In order to exclude such a situation, one-by-one comparisons must be performed between objects, and this is usually for all attributes--not just for position. This is very computationally expensive, because the complexity for such an operation is quadratic. As the number of objects and attributes increases, the computational cost for such an operation becomes too large very quickly and consumes too much time. In addition, a large number of single results must be combined into a final result, that is, one of the results must be either selected as a representative, or an average element must be calculated.

[0013] In this context, the use of a coordinate representation of the data set using geometric methods is proposed. For this purpose, the data set itself, such as is obtained from the original data set within the scope of the mentioned processing, should first be defined as a data set with one or more attributes. An attribute should be a value that should be included in the comparison.

[0014] In the case of (infrastructure-assisted) autonomous driving or in general in the case of vehicles, the attributes can, for example, each include information about a traffic participant, in particular a vehicle or an object on the road. The information about a traffic participant can, for example, include at least one of the following information: the position of the traffic participant, the orientation of the traffic participant, the speed of the traffic participant, the geometric dimensions of the traffic participant.

[0015] Thus, specific examples of attributes would be the global position in geographic longitude and latitude (which would be two attributes both related to, for example, the center of the vehicle or, for example, the center of gravity), the width of the vehicle, the length of the vehicle, and the orientation or direction of travel of the vehicle. The speed of the vehicle could also be such an attribute.

[0016] Thus, one of the multiple data sets where there may be due to the multiple processing paths represents an n-tuple (where n is equal to or greater than 1). Typically, each of the multiple data sets thus has the same number and type of attributes, but in principle the number and / or type of attributes may also differ from one another (e.g. because the processing paths use different types of calculations or because only certain parts are calculated).

[0017] In the best case, the corresponding attributes of the multiple data sets will be consistent and can be clearly assigned to each other, and their n-tuples will be the same, and therefore one of the attributes can be submitted as the final result. In most cases, the data sets will differ in the number of attributes and / or in one or more attributes. This may depend on the application or on whether data sets with a different number of attributes must already be considered different or can or should nevertheless participate in the method to be described below.

[0018] In order to compare a plurality of data sets, a comparison data set is provided for each of the plurality of data sets, which characterizes the corresponding data set. Each comparison data set has one or more comparison attributes. The comparison data set can correspond to the data set, so that the comparison attributes also correspond to the attributes (in terms of type and quantity). However, the comparison data set can also be derived (i.e. calculated) from the corresponding data set.

[0019] In the case of the already mentioned example of a vehicle and the mentioned attributes, the comparison attributes may be, for example, the global position (in 2D) of the geographical longitude and latitude for each of the four corner points of the vehicle (this would be eight comparison attributes). These geographical longitudes and latitudes for each of the four corner points can simply be determined or calculated based on the mentioned attributes and are a more intuitive representation. For example, speed would not be considered here.

[0020] It is also conceivable that some comparison properties correspond directly to the attribute, but other comparison properties are derived from the attribute.

[0021] When using a predefined common coordinate representation, the following then takes place for each of the plurality of comparison data sets. For the coordinate representation, for example, a coordinate system, for example a Cartesian coordinate system, can be used.

[0022] The coordinates of the coordinate representation are respectively assigned to one or each of a plurality of comparison attributes; it is also possible to speak of assigning dimensions of a multidimensional space. In the example of a Cartesian coordinate system, the coordinates would be, for example, coordinate axes, in 2D, for example the x-axis and the y-axis. However, in general, this may be an n-dimensional (Cartesian) coordinate system.

[0023] For the case where the comparison data set corresponds to the data set, and in the mentioned example and the mentioned properties of the vehicle, for example, the geographical longitude can be assigned to the coordinates (or coordinate axes), and likewise the geographical latitude can be assigned to the coordinates (or coordinate axes), and likewise the width of the vehicle, the length of the vehicle and the orientation or driving direction of the vehicle can be assigned to the coordinates (or coordinate axes). Here, this is a 5-tuple.

[0024] Based on the value of the corresponding comparison attribute, a corresponding coordinate value is then determined for the corresponding coordinate assigned to the comparison attribute. In the example mentioned, this could be, for example, a value for geographical longitude, a value for geographical latitude, a specific value for the width of the vehicle, etc.

[0025] However, it is not necessary to assign different coordinates (or coordinate axes) to each attribute or comparison attribute. Likewise, the same coordinates (or coordinate axes) can be assigned to two or more attributes or comparison attributes. In particular, the same coordinates (or coordinate axes) can also be assigned to a plurality of pairs or tuples (i.e., each having two or more attributes or comparison attributes). It is also possible to combine two or more attributes or comparison attributes, or, if necessary, to combine or calculate two or more attributes into one comparison attribute, which can then be assigned coordinates.

[0026] In the case where the comparison data set has been derived from the data set, and in the mentioned example of the vehicle and the mentioned comparison properties, for example, a Cartesian 2D coordinate system can be used as the coordinate representation. For example, the first coordinate axis can then be assigned to the geographical longitude of each of the four corner points, and the second coordinate axis can be assigned to the geographical latitude of each of the four corner points.

[0027] This results in representing the comparison dataset as a rectangle in a 2D coordinate system. Whereas in the case of using the dataset directly as the comparison dataset, this may correspond to representing the comparison dataset as a point in a 5D coordinate system.

[0028] In general, the comparison attributes can be appropriately assigned to the coordinates. If, for example, there are different, but partially comparable types of attributes or comparison attributes, such as temperature and an error range for the temperature and speed and an error range for the speed, for example, the temperature and the error range for the temperature can be assigned to a first coordinate axis, and the speed and the error range for the speed can be assigned to a second coordinate axis.

[0029] Now determine the common convex envelope of all coordinate values ​​of multiple comparison data sets. For the case where the comparison data set is represented as a rectangle in 2D, such a convex envelope will be a polygon, for example. However, for the case where the comparison data set is represented as a point in 5D, such a convex envelope will be a shape that cannot be represented graphically. It should be mentioned at this point that, depending on the number of coordinates and the way in which the comparison attributes are assigned to the coordinates, especially in terms of dimension, such an envelope may look different. In particular, the envelope should also be understood as an envelope that degenerates into a line. In particular, the convex envelope minimizes its surface (the surface may degenerate into a circle if necessary), that is, it just still surrounds points, etc.

[0030] The volume of the convex envelope is then determined. The volume can also be degenerated into an area or even a length here if necessary. Based on the comparison of the volume of the convex envelope with the comparison volume, it is then determined whether a plurality of data sets are assumed to be identical or to be identical within a predetermined tolerance. This is then provided as a comparison result for further use.

[0031] For example, if the volume of the convex envelope deviates from the comparison volume by less than a predetermined absolute or relative threshold, it can be assumed that the multiple data sets are identical or identical within a predetermined tolerance. The comparison volume can, in turn, correspond, for example, to the minimum volume, the maximum volume or the average volume (for example in the sense of an arithmetic mean) of a plurality of single volumes, wherein the plurality of single volumes respectively correspond to the volumes of the individual convex envelopes of the coordinate values ​​of the corresponding comparison data sets in the plurality of comparison data sets. Thus, if the volume of the common convex envelope deviates from the minimum single volume by at most 5%, it can be assumed, for example, that the data sets are identical within a predetermined tolerance.

[0032] However, instead of a comparison volume determined from a single volume, it is also possible to use values ​​that are assumed to be acceptable in other respects, for example, in the case of the vehicles mentioned and the representation of the comparison data set as rectangles, values ​​of typical volumes (or areas if necessary) of vehicles with tolerances. It is also possible to distinguish upward or downward deviations of the volume of the common convex envelope from the comparison volume.

[0033] For example, based on the comparison result, a final data set can then be determined for further use, such as operation of an autonomous vehicle, and based on a plurality of data sets. Thus, for example, one of the plurality of data sets or one of the plurality of data sets (e.g. an arithmetic mean, if necessary for each attribute) can be used as the final data set.

[0034] Depending on the type of application, volume calculations in n-dimensional or generally multidimensional space can also be approximated by a collection of small cubes. Depending on the type of attribute, it may be expedient to derive a comparison attribute such that the units are made comparable to one another.

[0035] With the proposed approach, it is now possible to speed up the comparator (or the comparison algorithm), for example to use it in a lockstep solution. In this case, a large number of object and attribute comparisons are not required, and instead an algorithm is used that finds out more quickly whether the provided single results (data sets with attributes) have sufficient similarity so that they can be regarded as unambiguous (correct) results and can be processed further. In addition, it is thus easier for the comparator to determine the final result without causing too much delay in the calculation for this purpose.

[0036] The computing system according to the invention, for example a computer center or a computer network having, for example, a plurality of computer centers, is configured, in particular in terms of programming, to carry out the method according to the invention.

[0037] It is also advantageous to implement the method according to the invention in the form of a computer program or computer program product with a program code for executing all method steps, because this causes particularly low costs, especially if the execution control device is also used for other tasks and therefore originally exists. Finally, a machine-readable storage medium is provided, with a computer program as described above stored thereon. Suitable storage media or data carriers for providing computer programs are especially magnetic, optical and electrical memories, such as hard disks, flash memories, EEPROMs, DVDs, etc. It is also possible to download programs via computer networks (the Internet, intranets, etc.). Here, such downloading can be done wired or cabled or wirelessly (e.g., via WLAN networks, 3G, 4G, 5G or 6G connections, etc.).

[0038] Further advantages and embodiments of the invention are apparent from the description and the accompanying drawings.

[0039] The invention is schematically illustrated in the drawings on the basis of exemplary embodiments and is described below with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 An environment for explaining the present invention is schematically shown.

[0041] Figure 2 The sequence of the method according to the invention in a preferred specific embodiment is schematically shown as a block diagram.

[0042] Figure 3a to Figure 3c A schematic diagram for explaining the present invention is shown.

[0043] Figure 4 A diagram for explaining the present invention is schematically shown. DETAILED DESCRIPTION

[0044] exist Figure 1 Schematically, an environment for explaining the present invention is shown in FIG. 1 , namely, by way of example, in the environment of so-called infrastructure-assisted autonomous driving.

[0045] For this purpose, a vehicle 100 is shown by way of example, which has a computing unit or computing system 102 and a sensor 104, which is designed, for example, as a camera. For example, calculations are performed with the aid of the computing unit or computing system 102, and a wireless communication connection can also be established, for example via an integrated radio module, in order to send and / or receive data. For example, objects in the environment can be detected with the aid of the camera 104.

[0046] Furthermore, an infrastructure object 110, such as a light signaling device, is shown as an example, which also has a computing unit or computing system 112 and a sensor 114, which is designed, for example, as a camera. With the aid of the computing unit or computing system 102, calculations can be performed, for example, and a wireless communication connection can be established, for example, via an integrated radio module, in order to send and / or receive data. For example, objects in the environment, such as a vehicle 100, can be detected with the aid of the camera 114.

[0047] Furthermore, an (external) computing system 150 is shown by way of example, by means of which, for example, calculations are performed, in particular, for example, also the method according to the invention is performed. A wireless communication connection can also be established, for example, via a radio module (for example a connected transmitting mast) 152, in order to send and / or receive data.

[0048] Within the scope of infrastructure-assisted autonomous driving, vehicle 100 can now be detected, for example, by means of camera 114 of traffic infrastructure object 110 and the sensor data generated thereby can be transmitted to computing system 150. There, the sensor data are processed in the sense of raw data sets into result data or generally data sets, which can then be used again, for example, for navigation of vehicle 100.

[0049] In order to achieve a certain security for the data or their correctness, the processing of the original data set can be carried out several times, for example three times, that is, in three different processing paths. The data sets obtained in this way can then be compared in order to ensure that the processing has been carried out in accordance with the regulations. This will be explained in more detail below.

[0050] exist Figure 2 The process of the method according to the present invention is schematically shown in a preferred embodiment. Figures 3a to 3c and 4 and also Figure 1 Explain this in more detail.

[0051] In step 200, for example, Figure 1 The camera 114 of the infrastructure object 110 detects the image of the vehicle 100. In this case, sensor data, also referred to below as raw data set 202, is obtained. This can be transmitted, for example, to a Figure 1The computing system 150 is then to be used for calculating the raw data in different, for example three different processing paths 206.1, 206.2, 206.3, step 204. In each of the processing paths 206.1, 206.2, 206.3, a data set is obtained, which is denoted here by 208.1, 208.2, 208.3. For example, position information etc. is determined from the images and possibly other information of the raw data set.

[0052] Each of the plurality of, here three, data sets 208.1, 208.2, 208.3 has one or more attributes. Typically, each data set has the same number and type of attributes, but this may also be different if necessary. In the figure, this is illustrated by way of example with attributes 210.1, 210.2, 210.3, which represent a plurality of attributes per data set.

[0053] exist Figure 3a This is illustrated graphically in . There, five attributes are shown by way of example for one of the data sets, namely by way of example in a 2D Cartesian coordinate system with axes x, y. Here, the attributes are the longitude 301 of the position (e.g. of the center of gravity of the vehicle), the latitude 302 of the position (301, 302 are represented by points in the coordinate system), the length 303 of the vehicle, the width 304 of the vehicle and the orientation (or driving direction) 305 of the vehicle.

[0054] Furthermore, by way of example, corresponding comparison data sets 214 . 1 , 214 . 2 , 214 . 3 with comparison properties indicated here by 216 . 1 , 216 . 2 , 216 . 3 are now to be determined or derived from data sets 208 . 1 , 208 . 2 , 208 . 3 with their properties (step 212 ).

[0055] This is Figure 3a 300. This is also illustrated graphically in FIG. There, for the data set with attributes that have already been shown, a comparative data set with comparative attributes is shown. In this case, the comparative attributes are the longitude 311x, 312x, 313x, 314x and the respectively associated latitude 311y, 312y, 313y, 314y for each of the four points. These are also represented in the coordinate system 300. These comparative attributes can be determined or calculated very easily from the attributes and can also be represented graphically in a simple manner. The points are the corner points (two-dimensional) of the vehicle, which is assumed here as a rectangle by way of example.

[0056] The coordinates of the coordinate representation are assigned to each comparison attribute, step 218 , and the coordinate values ​​are then respectively determined based on the values ​​of the corresponding comparison attributes, step 220 .

[0057] This is already Figure 3a , that is, using a 2D Cartesian coordinate system 300 with axes x, y as coordinate representation. In this case, the coordinate or axis x is respectively assigned to the longitude, and the coordinate or axis y is assigned to the latitude. The corresponding values ​​result from the positions shown.

[0058] These points each form a rectangle, for which a separate convex envelope 310 corresponding exactly to this rectangle can be determined.

[0059] exist Figure 3b This is shown for three comparative data sets in FIG. 3 , where the data sets or points are not explicitly labeled, but individual convex envelopes 310 , 320 , 330 are labeled. However, it can already be seen from this that the individual values ​​differ slightly from one another, although this is omitted for better illustration. Figure 3b are greatly exaggerated compared to their typical true values.

[0060] Then, a common convex envelope is determined for all coordinate values ​​of the comparison data set, step 222. Such a common convex envelope is Figure 3c denoted by 350. In this example this is the polygon defined by the points (321x, 321y), (322x, 322y), (332x, 332y), (333x, 333y), (334x, 334y), (324x, 324y). Some of the comparison attributes corresponding to the points are also within the common convex envelope 350.

[0061] In step 224, the volume of the common convex envelope is then determined. This volume is Figure 3c denoted by 352, and in this specific example has degenerated into area. Instead of volume, one can also talk about content.

[0062] The data set can also be used directly as a comparison data set, in which case, for example, each attribute or comparison attribute is also assigned a separate coordinate. Figure 4 In , this is shown by way of example with three data sets each having two attributes, and also with a 2D Cartesian coordinate system 400 with axes x, y as coordinate representation. Here, by way of example, there are attributes 401x, 401y, 402x, 402y, 403x, 403y. In this case, each data set represents a point in the coordinate system 400, whereby the individual convex envelopes degenerate into the point itself. The common convex envelope, which is here denoted by 410, corresponds in this example to a triangle whose volume has degenerated into an area.

[0063] In step 226, it is then determined whether the multiple data sets are assumed to be identical or identical within a predetermined tolerance. To this end, the volume of the convex envelope is compared with a comparison volume. The comparison volume can be, for example, the smallest single volume among the single volumes defined by the individual convex envelopes. Thereby a comparison result 228 is obtained, which can be provided, for example, in step 230.

[0064] For example, it can be provided that in the case of a deviation of at most 5% or another value, the data sets can be assumed to be identical within a predefined tolerance. The comparison result can then include the information that the data sets can be assumed to be identical within a predefined tolerance.

[0065] Based on the comparison result, for example, a final data set can then be determined for further use, for example for operating the autonomous vehicle, and based on the plurality of data sets, step 332. Thus, for example, one of the plurality of data sets or one of the plurality of data sets (for example the arithmetic mean, if necessary for each attribute) can be used as the final data set.

[0066] The proposed method is advantageous compared to a one-by-one comparison of all datasets because it can be used at a low cost. OO＝n logn (where n corresponds to the number of attributes or compared attributes) to achieve the computation of the convex envelope using the corresponding algorithm, while in other cases the cost would be quadratic.

Claims

1. A method for comparing a plurality of data sets (208.1, 208.2, 208.3), wherein each of the plurality of data sets (208.1, 208.2, 208.3) comprises one or more attributes (210.1, 210.2, 210.3), the method comprising: Providing, for each of the plurality of data sets (208.1, 208.2, 208.3), a comparison data set (214.1, 214.2, 214.3) characterizing the corresponding data set, wherein each comparison data set has one or more comparison attributes (216.1, 216.2, 216.3); For each of the plurality of comparison data sets, using a predetermined common coordinate representation: - assigning (218) the coordinates (x, y) of the coordinate representation (300, 400) to one or each of the plurality of comparison attributes, respectively, and - determining (220) a respective coordinate value assigned to the respective coordinate of the comparison attribute based on the value of the respective comparison attribute; determining (222) a common convex envelope (350) of all coordinate values ​​of the plurality of comparison data sets; determining (224) the volume (352) of the convex envelope; determining (226) based on a comparison of the volume of the convex envelope with a comparison volume: whether the plurality of data sets are assumed to be identical or to be identical within a predetermined tolerance as a result of the comparison (228); and The comparison result is provided (230) for further use.

2. The method according to claim 1, wherein if the volume (352) of the convex envelope deviates from the comparison volume by less than a predetermined absolute or relative threshold, it is assumed that the multiple data sets are identical or identical within a predetermined tolerance.

3. The method according to claim 2, wherein the comparison volume corresponds to a minimum volume, a maximum volume or an average volume among a plurality of single volumes, wherein the plurality of single volumes respectively correspond to the volumes of separate convex envelopes of the coordinate values ​​of corresponding comparison data sets in a plurality of comparison data sets.

4. The method according to claim 1 or 2, wherein for at least one of the plurality of comparison data sets (214.1, 214.2, 214.3), at least two of the plurality of comparison attributes are assigned to the same coordinate of the coordinate representation.

5. A method according to any of the preceding claims, wherein at least one of the multiple comparison data sets (214.1, 214.2, 214.3) corresponds to a corresponding data set, and / or wherein at least one of the multiple comparison data sets has been derived from a corresponding data set.

6. A method according to any of the preceding claims, wherein each of the multiple data sets (208.1, 208.2, 208.3) has been determined from a common original data set (202) by means of processing via corresponding processing paths (206.1, 206.2, 206.3), wherein the processing paths are different from each other.

7. The method according to claim 6, wherein the raw data set (202) has been acquired at least partially by means of one or more sensors (104, 114) of a vehicle (110) or a traffic infrastructure object (110). 8 . The method according to claim 6 , wherein the attributes of the plurality of data sets each include information about a traffic participant, in particular a vehicle or an object on the road. 9 . The method according to claim 8 , wherein the information on the traffic participant comprises at least one of the following information: the position of the traffic participant, the orientation of the traffic participant, the speed of the traffic participant, and the geometric dimensions of the traffic participant.

10. A computing system (150) which is configured to carry out all method steps of the method according to any one of the preceding claims.

11. A computer program which, when executed on a computing system (150), causes the computing system to perform all method steps of the method according to any one of claims 1 to 7.

12. A machine-readable storage medium having stored thereon the computer program according to claim 9.

Images