DE102019210128A1 - Method for evaluating recorded sensor data from a sensor - Google Patents

Abstract

The invention relates to a method for evaluating recorded sensor data (20) of a sensor (10), the recorded sensor data (20) being spatially resolved and including objects detected in an environment, the detected sensor data (20) of the sensor (10) also being spatially resolved Reference data (21) of a reference sensor (11) recorded in the environment are compared, and a sensor quality (30) of the sensor (10) is determined from a comparison result and output as an evaluation result (31). The invention also relates to a device (1) for data processing, a computer program and a computer-readable storage medium.

Description

The invention relates to a method for evaluating acquired sensor data from a sensor. The invention also relates to a device for data processing, a computer program and a computer-readable storage medium.

Modern motor vehicles are equipped with a large number of assistance systems which support a driver in driving the vehicle or which can drive the motor vehicle in a partially or automated manner. For this purpose, the assistance systems are supplied with a large number of sensor data from sensors (radar, LIDAR, ultrasound, infrared, etc.) that detect the surroundings of the motor vehicle. In order to provide functions, it is desirable to know how good a sensor quality of different types of sensors is in order to be able to make a selection of a suitable sensor for a function or an assistance system based on this.

The invention is based on the object of improving a method for evaluating acquired sensor data of a sensor.

According to the invention, the object is achieved by a method having the features of claim 1 and a device having the features of claim 8. Furthermore, the object is achieved according to the invention by a computer program with the features of patent claim 9 and a computer-readable storage medium with the features of patent claim 10. Advantageous embodiments of the invention emerge from the subclaims.

In particular, a method for evaluating acquired sensor data of a sensor is made available, the acquired sensor data being spatially resolved and comprising objects acquired in an environment, the acquired sensor data of the sensor being compared with reference data of a reference sensor acquired in the environment in a spatially resolved manner, and from A comparison result determines a sensor quality of the sensor and is output as an evaluation result.

Furthermore, in particular a device for data processing is created, comprising means for carrying out the method steps of the method according to the invention.

Furthermore, in particular a computer program is created, comprising commands which, when the computer program is executed by a computer, cause the computer to carry out the method steps of the method according to the invention.

Finally, in particular, a computer-readable storage medium is created, comprising instructions which, when executed by a computer, cause the computer to carry out the method steps of the method according to the invention.

The method and the device make it possible to evaluate a sensor with the aid of acquired sensor data. This is done through a comparison with reference data from a reference sensor recorded in the same environment. The reference sensor is in particular a sensor which has a higher sensor quality, in particular with regard to a resolution and a detection capability of objects in the environment. For example, a high-resolution LIDAR sensor can be used as a reference sensor to evaluate a radar or an ultrasonic sensor. In the example mentioned, the sensor data include spatially resolved radar or ultrasonic echoes from objects in the vicinity. Accordingly, the reference data includes LIDAR echoes from the objects in the environment. The sensor data and the reference data are in particular available as spatially resolved point clouds formed from measured values. Provision can be made for the reference data to be generated by multiple acquisition and subsequent averaging. The comparison of the sensor data recorded in the same environment of the sensor to be evaluated and the reference data takes place in an automated manner in particular. The result of the process is a certain sensor quality that enables a comparison with other sensors.

In particular, it is provided that the method is also carried out for at least one further sensor, the at least one further sensor using a different measuring method than the sensor and / or having different properties than this. Since a comparison with the same reference data is also carried out for the at least one further sensor, the respective sensor qualities determined for the sensor and the at least one further sensor can be compared directly with one another. In particular, a sensor recommendation and / or a sensor selection for a specific application in an assistance system of a motor vehicle can be given or made on the basis of the determined sensor qualities.

The recorded sensor data and the reference data are present in a spatially resolved manner, with individual measured values of the sensor data and the reference data relating in particular to the same coordinate system. The coordinate system is based, for example, on global coordinates or local Cartesian coordinates. The sensor data can be assigned to two-dimensional coordinates, which are additionally provided with height information. The height information here includes, in particular, information as to whether or not an object linked to a recorded measured value can be driven over or under by a motor vehicle in terms of height. A motorway bridge, for example, can be driven under, but not a barrier. Alternatively, the sensor data can also be assigned to three-dimensional coordinates. The reference data are assigned, in particular, to three-dimensional coordinates.

In particular, it can be provided that within the scope of the method several quality criteria for the sensor quality are determined and then weighted or unweighted are taken into account or combined to determine the sensor quality. It can also be provided that a statistical distribution of individual quality criteria, which were each determined in the same way for parts of the sensor data, is evaluated. A single quality criterion and / or an overall result can be standardized. It can be provided that a dependency of the overall result (s) on a selected quality criterion is determined and provided.

The method is carried out in particular as a computer-implemented invention.

In particular, the method is carried out by means of a computing device that can access a memory. The computing device can be designed as a combination of hardware and software, for example as program code that is executed on a microcontroller or microprocessor.

In one embodiment it is provided that to determine the sensor quality at least one polygon is generated on the basis of the spatially resolved reference data, the comparison of the acquired spatially resolved sensor data with the spatially resolved reference data being carried out on the basis of the at least one polygon. In this way, an outer contour of objects can be generated, which facilitates a comparison of the recorded sensor data with the reference data. The polygon course is generated, for example, by connecting measured values or associated points of the reference data with their respective closest neighbors.

In a further developing embodiment it is provided that a tolerance range is defined around the at least one polygonal line, with a number of measured values of the spatially resolved sensor data being determined to determine the sensor quality, which is within or outside the tolerance range. The sensor quality can then be determined, for example, on the basis of a ratio between the number within and the number outside the tolerance range. The sensor quality increases the more measured values are within the tolerance range.

It can be provided that the sensor quality is determined taking into account a standardization specification, the standardization specification using a portion of measured values that are within the tolerance range for a standardization. For example, it can be provided that at least 80% of the measured values must lie within the tolerance range so that a sensor quality is rated with a value of 1. If more measured values are within the tolerance range, the sensor quality is greater than 1, if fewer measured values are within the tolerance range, the sensor quality is less than 1. The standardization specification can alternatively or additionally include, for example, a number of measuring points per unit length or area of the polygon.

It can also be provided that a false detection rate is determined. The false detection rate here is a number of false detections, that is to say a number of measured values that are not in the tolerance range, divided by a total number of measured values. The false detection rate can then be taken into account as a quality criterion when determining the sensor quality.

It can further be provided that a point density is determined from a number of measured values in the sensor data per unit of length or unit of area within the tolerance range around the polygon. The point density can then be taken into account as a quality criterion when determining the sensor quality.

It can be provided that an error classification is also carried out and taken into account when determining the sensor quality. For example, it can be provided that an error classification of a measured value of the sensor data that is not in the tolerance range is dependent on an excess or underpassability occurs. If a measured value that is not within the tolerance range is not classified as being driven over or under, this measured value is assigned an error class “red”, for example. If, on the other hand, the measured value is classified as being driven over or under, this measured value is assigned the error class “yellow”. The respectively assigned error classes can then be used when determining the sensor quality to weight an associated quality criterion.

In one embodiment it is provided that measured value-wise a distance measure between measured values of the recorded sensor data and the at least one polygon is determined, for which purpose the measured values of the recorded sensor data are projected onto the at least one polygon, the sensor quality being determined on the basis of the respectively determined distance measures becomes. In order to determine the sensor quality, for example, a sum of a mean square distance between measured values and the respective associated projections can be calculated. The sum of the mean square distance can then be taken into account as a quality criterion when determining the sensor quality.

In a further-developing embodiment, it is provided that a statistical evaluation of the determined distance measures is carried out in order to determine the sensor quality. For example, an offset between measured values and a scatter of the measured values can be determined and taken into account as quality criteria when determining the sensor quality.

In one embodiment, it is provided that corners of objects are determined in the reference data, wherein to determine the sensor quality for each corner determined in the reference data, it is determined how many measured values of the detected sensor data are within a tolerance range around the corner. Since corners are an essential feature when detecting and recognizing objects, this allows the sensor to be assessed taking into account particularly relevant features of the objects.

In one embodiment it is provided that the determination of the sensor quality is selected as a function of a type and / or requirements of an assistance function for a motor vehicle. In particular, the quality criteria used can be selected on the basis of the type and / or the requirements of the assistance function. Furthermore, weightings that are used when determining and / or combining individual quality criteria can also be selected on the basis of the type and / or the requirements of the assistance function. For example, the tolerance range already described can be selected on the basis of the type and / or the requirements of the assistance function. A lane departure warning system can detect a guardrail with a greater tolerance than a parking assistant detects a post or a curb, for example. Correspondingly, a larger tolerance range, e.g. 50 cm, is chosen for evaluating a sensor required for the lane departure warning system, and a smaller tolerance range, e.g. 10 cm, is chosen for evaluating a sensor required for the parking assistant. It can also be provided that the tolerance range around the corners is selected as a function of a type and / or requirements of an assistance function for a motor vehicle.

In a further embodiment, it is provided that, in order to determine the sensor quality, a time profile of time-dependent sensor data recorded is taken into account and evaluated.

In one embodiment it is provided that objects are recognized and / or marked in the reference data, wherein when determining the sensor quality a weighting of acquired sensor data belonging to objects takes place on the basis of an object class assigned to the recognized and / or marked objects. In this way, for example, incorrect detections when detecting pedestrians can be considered more weighted than incorrect detections when detecting static objects such as trees or walls etc.

• 1 eine schematische Darstellung einer Ausführungsform der Vorrichtung zum Bewerten von erfassten Sensordaten eines Sensors;
• 2 eine schematische Darstellung von erfassten Sensordaten und einem aus Referenzdaten gebildeten Polygonzug zur Verdeutlichung einer Ausführungsform des Verfahrens;
• 3 eine schematische Darstellung von Messwerten in erfassten Sensordaten und einem aus Referenzdaten gebildeten Polygonzug zur Verdeutlichung einer Ausführungsform des Verfahrens;
• 4a eine schematische Darstellung von Messwerten und einem Polygonzug zur Verdeutlichung eines Versatzes von Messwerten;
• 4b eine schematische Darstellung von Messwerten und einem Polygonzug zur Verdeutlichung einer Streuung von Messwerten;
• 5 eine schematische Darstellung von Messwerten in erfassten Sensordaten und einem aus Referenzdaten gebildeten Polygonzug zur Verdeutlichung einer Ausführungsform des Verfahrens;
• 6 ein schematisches Ablaufdiagramm einer Ausführungsform des Verfahrens zum Bewerten von erfassten Sensordaten eines Sensors.

The invention is explained in more detail below on the basis of preferred exemplary embodiments with reference to the figures. Here show:

• 1 a schematic representation of an embodiment of the device for evaluating acquired sensor data of a sensor;
• 2 a schematic representation of acquired sensor data and a polygon course formed from reference data to illustrate an embodiment of the method;
• 3 a schematic representation of measured values in acquired sensor data and a polygon course formed from reference data to illustrate an embodiment of the method;
• 4a a schematic representation of measured values and a polygon to illustrate an offset of measured values;
• 4b a schematic representation of measured values and a polygon to illustrate a spread of measured values;
• 5 a schematic representation of measured values in acquired sensor data and a polygon course formed from reference data to illustrate an embodiment of the method;
• 6th a schematic flow chart of an embodiment of the method for evaluating acquired sensor data of a sensor.

In 1 Figure 3 is a schematic representation of an apparatus 1 shown for performing the procedure. The device 1 includes funds 2 to carry out the procedure. The means 2 comprise a computing device 3 and a memory 4th . To carry out the method steps, the computing device 3 on the memory 4th access and perform arithmetic operations in this.

In the memory 4th are recorded sensor data 20th of a sensor 10 and related reference data 21st a reference sensor 11 filed. The sensor data 20th and the reference data 21st were from the sensor 10 or the reference sensor 11 detected and the device 1 for example supplied via a suitable interface (not shown).

The sensor 10 is for example an ultrasonic or a radar sensor. The reference sensor 11 is, for example, a high-resolution LIDAR sensor. The reference data 21st of the reference sensor 11 , for example the LIDAR sensor, can also have been recorded several times and then averaged.

The recorded sensor data 20th are spatially resolved and include, for example, spatially resolved ultrasonic echoes or radar echoes from objects detected in an environment. The reference data 21st are also spatially resolved and include, for example, spatially resolved LIDAR echoes from the objects in the environment.

The computing device 3 compares the recorded sensor data 20th of the sensor 10 the reference data 21st of the reference sensor 11 and determines a sensor quality from a comparison result 30th of the sensor 10 . The specific sensor quality 30th is then used by the computing device 3 as an evaluation result 31 issued. The specific sensor quality 30th or the evaluation result 31 are output in particular in the form of a digital data packet.

Will this for several different sensors 10 repeatedly, it can be based on the determined sensor ratings 30th or the evaluation results 31 a comparison between the several different sensors 10 can be carried out with a sensor that is optimal under given boundary conditions (e.g. resolution, detection probability, costs, etc.) 10 can be identified.

In 2 is a schematic representation of measured values 22nd of acquired sensor data and a polygon course formed from reference data 23 shown to illustrate an embodiment of the method. The polygon was formed by connecting the closest neighbors of measured values of the reference data.

It can be provided here that the reference data are present in the form of measured values in a three-dimensional coordinate system. Alternatively, it can also be provided that the reference data are in the form of measured values in a two-dimensional coordinate system, each of which is additionally assigned height information that includes information as to whether the respective measured values have a height that a motor vehicle can drive over or under Not.

Around the polygon 23 around becomes a tolerance range for determining the sensor quality 24 placed. This can be fixed, for example on both sides on the basis of a distance to the polygon 23 be defined by 10 cm. Alternatively, the tolerance range can be provided 24 to choose depending on an assistance function of the motor vehicle for which an optimal sensor is to be found.

A number of measured values are used to determine the sensor quality 22nd the sensor data determined that are inside or outside the tolerance range 24 lie.

• - Ein mittlerer quadratischer Abstand: $$\text{mittl}\text{.quadr}\text{.Abstand}=\frac{\text{1}}{\text{N}}{{\displaystyle {\sum }_{k=1}^N\left(x_k^S-x_k^R\right)}}^2$$
  
  wobei N eine Anzahl von Messwerten im Toleranzbereich 24 bezeichnet, x^S _k und x^R _k bezeichnen eine Position eines betrachteten Messwertes 22 der Sensordaten bzw. eine Position eines zu diesem Messwert 22 nächstliegenden Messwert der Referenzdaten. Der Ausdruck in der Klammer entspricht dann einem kürzesten Abstand zwischen diesen Messwerten.
• - Eine Punktdichte: $$\text{Punktdichte }=\text{ N / L}$$
  
  wobei N die Anzahl von Messwerten im Toleranzbereich 24 und L eine Länge des Polygonzuges 23 ist.

For the readings 22nd that are within tolerance 24 are, for example, the following quality criteria can be taken into account when determining the sensor quality:

• - A mean square distance: $$\text{average}\text{.quadr}\text{.Distance}=\frac{\text{1}}{\text{N}}{{\displaystyle {\sum }_{k=1}^N\left(x_k^{\mathrm{S.}}-x_k^{\mathrm{R.}}\right)}}^2$$
  
  where N is a number of measured values in the tolerance range 24 denotes, x ^S _k and x ^R _k denote a position of a measured value under consideration 22nd the sensor data or a position for this measured value 22nd closest measured value of the reference data. The expression in brackets then corresponds to the shortest distance between these measured values.
• - A point density: $$\text{Point density}=\text{N / L}$$
  
  where N is the number of measured values in the tolerance range 24 and L is a length of the polygon 23 is.

wobei N die Anzahl der Messwerte ist.For measured values 22nd that are outside the tolerance range 24 an error detection rate can be determined: $$\text{False detection rate}=\text{Number of false detections / N}$$

where N is the number of readings.

It can also be provided that an incorrectly detected measured value 25th , that is, a measured value 25th out of tolerance 24 , depending on whether the incorrectly detected measured value can be driven over or under 25th is provided with an error class. Is the reading 25th classified as being driven over or under due to an associated height, the incorrectly detected measured value 25th an error class "yellow" is assigned. Is the reading 25th on the other hand, classified as cannot be driven over or under, the incorrectly detected measured value 25th an error class "red" is assigned. Depending on the assigned error class, the incorrectly detected measured values can then be weighted to different degrees when determining the sensor quality 25th be made.

In 3 is a schematic representation of measured values 22nd in recorded sensor data and a polygon formed from reference data 23 shown to illustrate an embodiment of the method.

To determine the sensor quality or a quality criterion, a distance measure is provided, measured value by measurement Δk between readings 22nd the acquired sensor data and the at least one polygon 23 to be determined, for this purpose a projection 26th of the measured values 22nd the acquired sensor data on the polygon 23 takes place, the sensor quality or the quality criterion on the basis of the respective determined distance dimensions Δk is determined.

wobei N eine Anzahl von Messpunkten und L eine Länge des Polygonzuges bezeichnen. Mit einem Maximum der mittleren quadratischen Abweichung $$f_{max}={\left(\frac{L}{N+1}\right)}^2+\frac{1}{N}{\left(L-\frac{L}{N+1}\right)}^2$$

ergibt sich ein auf einen Bereich von [0,1] normiertes Gütefunktional: $$I=\frac{1}{f_{max}}f\left(\Delta \right)$$

A quality functional from a mean square deviation of the distance dimensions can be used as a quality criterion Δk be determined: $$f\left(\Delta \right)=\frac{\text{1}}{\text{N}}{{\displaystyle {\sum }_{k=1}^{N+1}\left({\Delta }_k-\frac{\mathrm{L.}}{N+1}\right)}}^2$$

where N denotes a number of measuring points and L denotes the length of the polygon. With a maximum of the mean square deviation $$f_{max}={\left(\frac{\mathrm{L.}}{N+1}\right)}^2+\frac{1}{N}{\left(\mathrm{L.}-\frac{\mathrm{L.}}{N+1}\right)}^2$$

the result is a quality functional normalized to a range of [0.1]: $$\mathrm{I.}=\frac{1}{f_{max}}f\left(\Delta \right)$$

It can be provided that the calculated quality functional is then statistically evaluated. For example, an offset and / or a spread of measured values 22nd to be determined. This is shown schematically in the 4a and 4b shown.

The effect caused by an offset is shown schematically in FIG 4a shown. Here are all measured values 22nd by an equal offset 27 versus the projections 26th on the traverse 23 postponed.

The effect caused by scattering is shown schematically in FIG 4b shown. Here are measured values 22nd around the projections 26th on the traverse 23 scattered around, whereby the scatter can be statistically evaluated and quantified.

A certain offset and a certain scatter can then be taken into account as a quality criterion when evaluating the sensor.

In 5 is a schematic representation of measured values 22nd in recorded sensor data and a polygon formed from reference data 23 shown to illustrate an embodiment of the method.

At the polygon 23 become corners 28 identified. This is done, for example, by changing a slope of the polygon 23 is evaluated. Places where an incline changes significantly or by a specified amount are called corners 28 Are defined. Around these corners 28 becomes a round tolerance range 29 Are defined.

It can be provided here that the round tolerance range 29 around the corners 28 around depending on a type and / or requirements of an assistance function for a motor vehicle is selected.

Then it is determined how many of the measured values 22nd the sensor data is within the round tolerance ranges 29 are located. It can also be determined whether all corners 28 were recognized, for example by checking whether at least one measured value 22nd within the tolerance ranges 22nd present.

A recognition rate, for example, can be calculated as a quality criterion: $$\text{Recognition rate}=\text{Number of corners recognized / number of corners}$$

wobei N die Anzahl der Ecken 28, x^s _k die Position des jeweiligen Messwertes 22 des Sensors innerhalb des runden Toleranzbereiches 29 und x^R _k die zugehörige Position der Ecke 28 auf dem Polygonzug 23 bezeichnen.It can also be used for recognized corners 28 a mean square distance is determined and for all corners 28 be summed up: $$\text{average}\text{.quadr}\text{.Distance}=\frac{\text{1}}{\text{N}}{{\displaystyle {\sum }_{k=1}^N\left(x_k^{\mathrm{S.}}-x_k^{\mathrm{R.}}\right)}}^2$$

where N is the number of corners 28 , x ^s _{k is} the position of the respective measured value 22nd of the sensor within the round tolerance range 29 and x ^R _{k is} the corresponding position of the corner 28 on the traverse 23 describe.

In 6th a schematic flowchart of an embodiment of the method for evaluating acquired sensor data of a sensor is shown.

After a start 100 the process is carried out in one process step 101 a sensor that has not yet been evaluated is selected from a list of sensors to be evaluated.

In one process step 102 sensor data of the selected sensor and the corresponding reference data, that is to say recorded in the same environment, are provided.

The recorded sensor data and the reference data are processed in one process step 103 compared to each other. The procedural step 103 comprises the procedural steps 104 to 109 .

In one process step 104 a polygon is formed by connecting neighboring measured values in the reference data.

In one process step 105 several quality criteria are determined for evaluating the acquired sensor data or the sensor. The procedural step 105 includes the procedural steps for this 106 to 108 . In the process step 106 a tolerance range is defined around the polygon and it is determined how many measured values of the sensor data are within this tolerance range. A false detection rate is determined from this. In the process step 107 is used for the measured values within the process step 106 defined tolerance range, determines how large a sum of a mean square distance between the measured values of the sensor and the polygon is. In one process step 108 it is determined how great a density (measured values per unit length / area / volume) of the measured values within the defined tolerance range is.

In one process step 109 the specific quality criteria are combined to form a sensor quality. This can be done, for example, by weighted summation. It can be provided here to normalize the quality criteria and / or the sensor quality.

In one process step 110 the determined sensor quality is provided and output as an evaluation result, for example in the form of a sensor quality signal.

In one process step 111 it is checked whether there are any further sensors to be evaluated in the list. If this is the case, becomes a process step 101 bounced back.

If, on the other hand, this is not the case, the provided sensor qualities of the individual sensors are determined in one process step 111 compared with each other and identified one of the sensors in the list as the optimal sensor based on a comparison result. This can be done taking into account other boundary conditions, such as costs or installation space requirements, etc. The selection result is then made in one process step 113 issued.

The procedure is then ended 114 .

It can be provided that the determination of the quality criteria is selected as a function of a type and / or requirements of an assistance function for a motor vehicle for which an optimal sensor is to be found. In particular, for example, in process step 106 defined tolerance range can be selected depending on the type and / or the requirements of the assistance function for the motor vehicle.

List of reference symbols

1

contraption

2

medium

3

Computing device

4th

Storage

10

sensor

11

Reference sensor

20th

Sensor data

21st

Reference data

22nd

Measured value

23

Traverse

24

Tolerance range

25th

incorrectly detected measured value

26th

projection

27

Offset

28

corner

29

round tolerance range

30th

Sensor quality

31

Evaluation result

100-114

Procedural steps

Δk

Spacing

Claims (10)

Method for evaluating captured sensor data (20) of a sensor (10), the captured sensor data (20) being spatially resolved and comprising objects captured in an environment, the captured sensor data (20) of the sensor (10) also being captured spatially resolved in the environment Reference data (21) of a reference sensor (11) are compared, and a sensor quality (30) of the sensor (10) is determined from a comparison result and output as an evaluation result (31). Procedure according to Claim 1 , characterized in that to determine the sensor quality (30) at least one polygon (23) is generated on the basis of the spatially resolved reference data (21), the comparison of the detected spatially resolved sensor data (20) with the spatially resolved reference data (21) on the basis of the at least a traverse (23) takes place. Procedure according to Claim 2 , characterized in that a tolerance range (24) is defined around the at least one polygonal course (23), with a number of measured values (22) of the spatially resolved sensor data (20) being determined to determine the sensor quality (30), which are inside or outside of the tolerance range (24). Procedure according to Claim 2 , characterized in that a distance (Δk) between measured values (22) of the acquired sensor data (20) and the at least one polygon course (23) is determined measured value-wise, a projection (26) of the measured values (22) of the acquired sensor data ( 20) takes place on the at least one polygonal course (23), the sensor quality (30) being determined on the basis of the distance dimensions (Δk) determined in each case. Procedure according to Claim 4 , characterized in that a statistical evaluation of the determined distance measures (Δk) is carried out to determine the sensor quality (30). Method according to one of the preceding claims, characterized in that corners (28) of objects are determined in the reference data (21), wherein for determining the sensor quality (30) for each corner (28) determined in the reference data (21) it is determined how many measured values (22) of the recorded sensor data (20) lie within a tolerance range (29) around the corner (28). Method according to one of the preceding claims, characterized in that the determination of the sensor quality is selected as a function of a type and / or requirements of an assistance function for a motor vehicle. Device (1) for data processing, comprising means (2) for carrying out the method steps of the method according to any one of the Claims 1 to 7th . Computer program, comprising instructions which, when the computer program is executed by a computer, cause the computer program to perform the method steps of the method according to any one of the Claims 1 to 7th execute. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the method steps of the method according to any of the Claims 1 to 7th execute.

Images