Method for operating a sensor device

Abstract

The invention relates to a method for operating a sensor device (100) for different vehicle functions outside and / or inside a vehicle (F) on the basis of radar technology, preferably UWB radar technology, from which radar measurements (CIR) provide time-of-flight information from which a point location (Tap) of a potential target (Z) can be derived, the continuous radar measurements (CIR) provide a change in time-of-flight information from which a movement, such as a speed, of a potential target (Z) at a point location (Tap) can be derived. The invention further relates to a corresponding computer program product, a control unit and a sensor device.

Description

TECHNICAL FIELD

[0001] The present application relates to a method for operating a sensor device based on radar technology, preferably ultra-wideband (UWB) radar technology. Furthermore, the present application relates to a corresponding computer program product, a corresponding control unit and a corresponding sensor device, in particular a UWB sensor device, for implementing the corresponding method. BACKGROUND

[0002] UWB sensor devices with radar functionality installed in a vehicle (typically one unit per seat row, but any number is possible) and outside the vehicle (typically one unit at each corner) can be used for different vehicle functions inside and / or outside the vehicle. Envisageable vehicle functions are, for example, the identification of intrusions, the contactless opening of vehicle flaps (e.g. doors or tailgates), the tracking of objects outside the vehicle (e.g. vehicles and / or persons), the identification of objects inside the vehicle (e.g. children and / or animals), etc. In general, UWB sensor devices can be used for the identification, localization, tracking and classification of passive objects (persons, animals, vehicles, etc.). These objects do not necessarily, but can communicate with the UWB sensor device.

[0003] UWB radar measurements can be described with respect to fast time. Here, a radar measurement can be evaluated which provides time-of-flight information in order to derive point positions (often referred to as taps) from the time-of-flight information. UWB radar measurements can be described with respect to slow time. Here, a plurality of consecutive radar measurements (channel impulse response or CIR for short) can be evaluated in order to derive a change of the signal at each point position. The analysis of the fast time can be used to determine point positions. The analysis of the slow time can be used to identify movements.

[0004] A standard method for the analysis of the slow time consists in converting the analysis into the frequency domain, for example by means of the so-called fast Fourier transform. In order to determine the dominant frequencies and thus the velocities accurately in the radar measurement, this analysis requires a high scanning frequency and a large number of data points, both of which can not be available in the case of UWB radar applications. Furthermore, the subsequent calculation operations are very complex in terms of numerical value. SUMMARY

[0005] It is therefore the task of the present application to at least partially overcome at least one of the above-mentioned disadvantages. In particular, it is the task of the present application to provide an improved method for operating a sensor device based on radar technology, preferably UWB radar technology. Preferably, it is the task of the present application to develop a simple, fast and efficient analysis of a motion in order to determine whether there is relevant information about a slow time at a determined place that has to be processed. Furthermore, it is the task of the present application to provide a corresponding computer program product, a corresponding control unit and a corresponding sensor device, in particular a UWB sensor device, for implementing a corresponding method.

[0006] The task of the present application is solved by a method having the features of independent method claim. Furthermore, the task of the present application is solved by a corresponding computer program product, a corresponding control unit and a corresponding sensor device, in particular a UWB sensor device, having the features of the dependent patent claims. Herein, the features and details described in connection with different embodiments and / or aspects of the present application apply of course also in connection with the other embodiments and / or aspects and vice versa, respectively, so that the disclosure with respect to each embodiment and / or aspect always refers to each other or can refer to each other.

[0007] The present application provides:

[0008] A method for operating a sensor device based on radar technology, preferably UWB radar technology, which can be used for different vehicle functions outside and / or inside a vehicle.

[0009] In a vehicle at least one or a plurality of sensor devices can be installed per seat row. Outside a vehicle at least one or a plurality of sensor devices can be installed at each corner and / or each side of the vehicle.

[0010] Possible vehicle functions are for example the recognition of an intrusion, the contactless opening of a vehicle flap (e.g. a door or a rear flap), the tracking of objects outside a vehicle (e.g. vehicles and / or persons), the recognition of objects inside a vehicle (e.g. children and / or animals) and the like.

[0011] A plurality of radar measurements can be implemented in succession by means of a sensor device. A radar measurement can comprise the emission of an emission signal (e.g. a UWB signal pulse) and / or the reception of a reception signal reflected for example on a potential target.

[0012] A recognized target can be an object. Possible objects outside a vehicle can be for example vehicles, bicycles, scooters, persons and the like. Possible objects inside a vehicle can be for example persons, children, animals and the like.

[0013] Radar measurements can provide time-of-flight information from which a point location of a potential target (often referred to as a Tap, where 1 Tap can be for example about 15 cm) can be derived.

[0014] Continuous radar measurements can provide a change in time-of-flight information from which a motion, for example a velocity, of a potential target at a point location can be derived.

[0015] The proposed method has here the following actions or in other words method steps:

[0016] - selecting a point location in one radar measurement,

[0017] - detecting continuous radar measurements for the determined point location.

[0018] - determining a significance value for the selected point location in the continuous radar measurements,

[0019] - evaluating the selected point location as significant or insignificant for further analysis, for example for determining a motion of a potential target at the selected point location, depending on the significance value.

[0020] The method can be used to quickly and simply determine whether there is relevant information for further analyzing the motion with respect to the slow time at a determined location. By means of the method, the amount of information or data for further analyzing the motion with respect to the slow time can advantageously be reduced to 10% of the total amount. In this way, a simple, quick and efficient analysis of the slow time can be achieved. Overall, an improved method for operating a sensor device can thereby be provided, which can be implemented with less computational complexity and reduced storage requirements.

[0021] The concept advantageously makes use of the following observations:

[0022] 1. Random noise is independent of time.

[0023] 2. When no relevant events occur at a determined location (or a determined Tap), the slow-time sequence of random noise dominates.

[0024] 3. When some relevant events occur at a determined location, the slow-time sequence shows a structure that is dependent on time.

[0025] The significance value can for example be defined in the sense of the present application as a function, for example as a ratio between a change of data values in the CIR (number n of CIRs within a determined time interval) and a change of differences of data values in the CIR (for example standard deviation, variance, etc.).

[0026] The significance value (English: "Significant Score" for SW) for a selected Tap can be expressed, for example, in the following formula:

[0027]

[0028] Herein:

[0029] CIR i is the CIR of the number i, wherein i = 1, 2,... to n, wherein n is the number of radar measurements,

[0030] CIR mean is the average value of the n number of CIRs,

[0031] ACIR i is the difference between two consecutive CIRs,

[0032] ACIR mean is the average value of the difference between two consecutive CIRs.

[0033] In order to assess whether a significant information is contained at a certain point (or Tap), the following assessment can be carried out.

[0034] When the typical signal change between two consecutive CIRs at a certain Tap in the UWB signal is as important as the typical signal change over a plurality of CIRs, then the Tap does not contain motion information.

[0035] When, however, the typical signal change between two consecutive CIRs at a certain Tap is greater than the typical signal change over a plurality of CIRs, then the Tap can contain important motion information.

[0036] It can furthermore be provided that the method can be carried out iteratively for different points, for example for all available Taps in the radar measurements. In this way different points can be assessed with respect to relevant motion information in order to select only relevant points and thus reduce the amount of information for further analysis.

[0037] It can furthermore be provided that the detection of consecutive radar measurements is carried out for the determined points until a certain number of consecutive radar measurements has been carried out. In this way a signal analysis with respect to the slow time can be achieved.

[0038] It can furthermore be provided that the significance value is formed in such a way that a distinction between the usual background noise in the consecutive radar measurements and a determined time sequence (which for example shows a determinable frequency) is possible. A distinction between the usual background noise and the relevant motion information can thus be made.

[0039] As already mentioned above, it is conceivable that the significance value can be determined as a ratio between a change of the data values in the continuous radar measurement and a change of the difference of the data values in the continuous radar measurement.

[0040] Advantageously, the method, in particular the method comprising the determination of the significance value and / or the evaluation of the selected tap as significant or insignificant for a further analysis, can be implemented by means of a method for machine learning. In this way, a computationally simple and storage-technically advantageous solution can be provided. Furthermore, a method for machine learning can reliably and efficiently identify time correlations and / or frequencies in the signal in order to avoid complex analysis by means of a so-called fast Fourier transform.

[0041] Preferably, the further analysis can be carried out at the preferably best, evaluated significant tap. In this way, only one or the best tap can thus be used for the motion analysis.

[0042] Advantageously, a maximum of four taps can be selected as significant for the further analysis. In this way, the amount of data for the further analysis can be reduced and, furthermore, it can be ensured that potential targets are reliably identified even in cases of uncertainty.

[0043] In principle, it can be advantageous for the method to select so many taps as significant that the amount of data for the further analysis is reduced to a maximum of 10% of the data obtained by means of the continuous radar measurement. In this way, a balanced solution for reducing the amount of data and reliably identifying potential targets can be provided.

[0044] Furthermore, it can be provided that a threshold value for the significance value is determined, according to which the selected taps are evaluated as significant or insignificant. In this way, a quick and simple evaluation of the selected taps can be achieved.

[0045] Furthermore, it can be provided that the position of the potential target relative to the vehicle, which can be located inside or outside the vehicle, is taken into account when evaluating the selected taps as significant or insignificant for a desired vehicle function inside and / or outside the vehicle. In this way, a simple additional criterion can be provided in order to evaluate whether a potential target can be important for a desired vehicle function. For example, it can be determined in the vehicle function "identify a child parked inside the vehicle" that potential targets outside the vehicle are ignored and not evaluated with respect to the slow time.

[0046] In principle, it can be advantageous that potential targets outside the vehicle can be ignored for vehicle functions inside the vehicle.

[0047] Furthermore, it can be provided that the method is implemented for tracking objects, such as vehicles and / or persons, outside and / or inside the vehicle.

[0048] It can be advantageous in principle for vehicle functions outside the vehicle to be able to ignore potential targets inside the vehicle.

[0049] It can furthermore be advantageous for measures to be introduced, for example for an external camera to be activated, for lighting to be switched on, for an alarm to be output, etc., when an object outside the vehicle is tracked and a false action of the object is recognized.

[0050] It can furthermore be advantageous for an identity check to be triggered when an object is tracked from outside the vehicle into the interior of the vehicle. In this way, reliable and comfortable keyless access control can be implemented.

[0051] Advantageously, the method can be implemented for operating a plurality of sensor devices. In this way, an extended analysis can be implemented.

[0052] It is conceivable here for the information of the plurality of sensor devices to be pooled in order to provide different vehicle functions outside and / or inside the vehicle with greater reliability and improved customer comfort.

[0053] The invention furthermore provides:

[0054] A computer program product, comprising instructions which, when the computer program product is executed by a computer, cause the computer to carry out a corresponding method, which can be carried out as described above. The same advantages as described above in connection with the method can be achieved here. These advantages are hereby fully incorporated.

[0055] The invention furthermore provides:

[0056] A control unit, having a storage unit in which a code is stored, wherein, when the code is executed by a computing unit, a corresponding method is carried out, which can be carried out as described above. The same advantages as described above in connection with the method can be achieved here. These advantages are hereby fully incorporated.

[0057] It is conceivable in particular for the control unit of one of the plurality of sensor devices to have a master function in order to pool the information of the plurality of sensor devices when the method is implemented for operating a plurality of sensor devices.

[0058] But it is also conceivable in principle for a plurality of control units to each have an evaluation function in order to pool the information of the plurality of sensor devices independently of one another.

[0059] The invention furthermore provides:

[0060] A sensor device, in particular a UWB sensor device, has a corresponding control unit, which is executed for implementing a corresponding method, which can be carried out as described above. The same advantages as described above in connection with the method can be achieved here. These advantages are fully referred to here. BRIEF DESCRIPTION OF DRAWINGS

[0061] The application is explained further below with the aid of the drawings. Herein are shown schematically, respectively:

[0062] Figure 1 An exemplary vehicle comprising a plurality of sensor devices is shown;

[0063] Figure 2 Exemplary radar measurements for a plurality of Taps are shown;

[0064] Figure 3 Exemplary radar measurements for one Tap are shown;

[0065] Figure 4 Exemplary saliency values for a plurality of radar measurements are shown, respectively for one Tap; and

[0066] Figure 5 An exemplary flow of the proposed method is shown. DETAILED DESCRIPTION

[0067] Figures 1 to 5 For explaining the proposed concept, the concept gives:

[0068] A method for operating a sensor device 100 based on radar technology, preferably UWB radar technology, which can be used for different vehicle functions outside and / or inside a vehicle F.

[0069] In the vehicle F at least one or a plurality of sensor devices 100 can be installed per seat row (cf. Figure 1 ). Outside the vehicle F at least one or a plurality of sensor devices 100 can likewise be installed on each corner and / or each side of the vehicle F.

[0070] Contemplatable vehicle functions that can be provided with the sensor device 100 are, for example, the recognition of intrusions, the contactless opening of vehicle flaps (for example doors or rear flaps), the tracking of objects (for example vehicles and / or people) outside the vehicle F, the recognition of objects (for example children and / or animals) inside the vehicle F, etc.

[0071] A plurality of radar measurements CIR can be carried out in turn with the sensor device 100 (cf. Figure 2 ). The radar measurement CIR can comprise the emission of an emission signal (for example a UWB signal pulse) and / or the reception of a reception signal (for example a reflected UWB signal pulse) reflected, for example, on a potential target.

[0072] The recognized object can be an object. Possible objects outside the vehicle can be, for example, vehicles, bicycles, scooters, people, etc. Possible objects inside the vehicle can be, for example, people, children, animals, etc.

[0073] The radar measurement CIR can provide time-of-flight information from which a point location (often referred to as Tap, where 1 Tap can be, for example, approximately 15 cm) of a potential object Z can be derived (cf. Figure 2 ).

[0074] The continuous radar measurement CIR can provide a change in the time-of-flight information from which a movement, for example a velocity, of a potential object Z at a point location Tap can be derived (cf. Figure 2 ).

[0075] As Figure 5 As indicated, the method can comprise the following method steps:

[0076] 110: selecting a point location Tap in a radar measurement CIR,

[0077] 120: detecting a continuous radar measurement CIR for the determined point location Tap,

[0078] 130: determining a significance value SW in the continuous radar measurement CIR for the selected point location Tap,

[0079] 140: evaluating the selected point location Tap as significant (true - cf. method step 141) or not significant (false - cf. method step 142) for further analysis, for example for determining a movement, for example a velocity, of a potential object Z at the selected point location Tap, depending on the significance value SW.

[0080] By means of the method, it can be quickly and simply determined whether relevant information about the slow time for further analysis of the movement can be obtained at a determined point location Tap. By means of the method, it can be advantageously reduced to 10% of the total amount the amount of information or data about the slow time for further analysis of the movement.

[0081] The following insight is considered here:

[0082] 1. Random noise is independent of time.

[0083] 2. When no relevant events occur at a determined location (or a determined Tap), the slow-time sequence of random noise dominates (cf. Figure 3 lower part).

[0084] 3. When certain relevant events occur at a certain point (Tap), the slow time series shows a structure related to time (cf. upper part of Figure 3 ).

[0085] The significance value (SW, cf. lower part of Figure 4 ) can in the sense of the present application for example be understood as a ratio of the variation (e.g. standard deviation, variance, etc.) of the data values in the CIR (number n of CIRs in a certain time interval) to the variation of the difference of the data values in the CIR.

[0086] In order to assess whether a certain point (or Tap) contains significant information or not, the following assessment can be carried out.

[0087] When the typical signal variation between two consecutive CIRs in a UWB signal at a certain Tap is as important as the typical signal variation over a plurality of CIRs, then this Tap does not contain motion information (cf. lower part of Figure 3 ).

[0088] When the typical signal variation between two consecutive CIRs in a UWB signal at a certain Tap is greater than the typical signal variation over a plurality of CIRs, then this Tap can contain important motion information (cf. upper part of Figure 3 ).

[0089] Advantageously, the method can be carried out iteratively for different points Tap, preferably for all available Taps in the radar measurement CIR, in order to select relevant Taps for further analysis.

[0090] In principle, n consecutive radar measurement CIRs can be carried out for a certain point Tap.

[0091] Advantageously, the significance value SW can be formed such that a distinction between the usual background noise in the consecutive radar measurement CIRs and a certain time series (which for example shows a determinable frequency) is possible.

[0092] As already mentioned above, the significance value SW can be determined depending on or as a ratio between the variation of the data values in the consecutive radar measurement CIRs and the variation of the difference of the data values in the consecutive radar measurement CIRs.

[0093] The significance value (in English "Significant Score" for SW) for a selected Tap can for example be expressed in the following formula:

[0094]

[0095] Here:

[0096] CIR i is the CIR of the i-th time, wherein i = 1, 2,... to n, wherein n is the number of radar measured CIRs,

[0097] CIR mean is the average of the n number of CIRs,

[0098] ACIR i is the difference between two consecutive CIRs,

[0099] ACIR mean is the average of the difference between two consecutive CIRs.

[0100] Advantageously, the method, in particular including determining a significance value (SW) and / or evaluating the selected tap (Tap) as significant (true) or not significant (false) for further analysis, can be implemented by means of a method for machine learning, so that time correlations and / or frequencies in the (UWB) signal can be reliably and efficiently identified.

[0101] For example, further analysis can be implemented only at the preferred best, evaluated as significant (true) taps Tap.

[0102] For example, a maximum of four taps Tap can be selected as significant for further analysis.

[0103] In principle, it can be advantageous for the method to select so many taps Tap as significant that the amount of data for further analysis is reduced to 10% of the data at all.

[0104] As Figure 4 Indicated, a threshold value SW* can be determined for the significance value SW, according to which the selected tap Tap is evaluated as significant (true) or not significant (false).

[0105] In the method, the position of the potential target Z relative to the vehicle F can be taken into account, which can be located inside or outside the vehicle F. In this way, for example, it can be determined in the vehicle function "identify a child inside a parked vehicle": ignore and do not evaluate the potential target outside the vehicle F with regard to the slow time.

[0106] In principle, potential targets outside the vehicle F can be ignored for vehicle functions inside the vehicle F.

[0107] Advantageously, the method can be implemented for tracking objects, for example vehicles and / or persons, outside and / or inside the vehicle F.

[0108] In principle, potential targets inside the vehicle F can be ignored for vehicle functions outside the vehicle F.

[0109] When an erroneous behavior of the object is recognized when tracking the object outside the vehicle F, then measures can be introduced, e.g. activating external cameras, switching on lighting, outputting an alarm, etc.

[0110] Furthermore, the method can be used for tracking an object when moving from outside the vehicle F into the interior space of the vehicle in order to introduce an identity verification, e.g. for a reliable and comfortable keyless access control.

[0111] The method can also be implemented for operating a plurality of sensor devices 100.

[0112] The information of the plurality of sensor devices 100 can be aggregated here in order to provide different vehicle functions outside and / or inside the vehicle with higher reliability and improved customer comfort.

[0113] A corresponding computer program product, a corresponding control unit ECU and a corresponding sensor device 100, in particular a UWB sensor device, constitute further aspects of the application.

[0114] The foregoing description of the drawings merely explains certain aspects of the application. Of course, the various features of the described embodiments can be freely combined with each other, as long as this is technically possible, without leaving the scope of the application.

[0115] List of reference signs

[0116] 100 sensor device

[0117] F vehicle

[0118] Z target

[0119] Tap point

[0120] CIR radar measurement

[0121] SW significance value

[0122] SW* threshold value

[0123] true significant

[0124] false insignificant

[0125] N number

[0126] ECU control unit

Claims

1. Method for operating a sensor device (100) based on radar technology, preferably UWB radar technology, for different vehicle functions outside and / or inside a vehicle (F), wherein - the radar measurement (CIR) provides time-of-flight information from which a point location (Tap) of a potential target (Z) can be derived, - wherein the successive radar measurements (CIR) provide changes in the time-of-flight information from which a movement, for example a velocity, of the potential target (Z) at one point location (Tap) can be derived, and wherein the method comprises: - selecting one point location (Tap) in one radar measurement (CIR), - detecting successive radar measurements (CIR) for the determined point location (Tap), - determining a significance value (SW) for the selected point location (Tap) in the successive radar measurements (CIR), - evaluating the selected point location (Tap) as significant (true) or not significant (false) for further analysis, for example for determining a movement of the potential target (Z) at this selected point location (Tap), depending on the significance value (SW).

2. Method according to claim 1, wherein - the method is implemented iteratively for different point locations (Tap), and / or wherein the detection of successive radar measurements (CIR) is implemented for the determined point location (Tap) until a determined number (n) of successive radar measurements (CIR) has been implemented.

3. Method according to claim 1 or 2, wherein - the significance value (SW) is formed in such a way that a distinction between usual background noise in the successive radar measurements (CIR) and a determined time sequence is possible, and / or wherein the significance value (SW) is determined depending on or as a ratio between a change in data values in the successive radar measurements (CIR) and a change in difference values of data values in the successive radar measurements (CIR).

4. Method according to one of the above claims, wherein - the method is implemented by means of a method for machine learning, in particular the method including the determination of the significance value (SW) and / or the evaluation of the selected point location (Tap) as significant (true) or not significant (false) for further analysis is implemented by means of a method for machine learning.

5. Method according to one of the above claims, wherein - the further analysis is implemented at the preferably best, evaluated as significant (true) point location (Tap).

6. Method according to one of the above claims, wherein - up to four point locations (Tap) are selected as significant (true) in order to implement the further analysis.

7. Method according to one of the above claims, wherein, - as many point locations (Tap) are selected as significant (true) by the method that the amount of data for the further analysis is reduced at all times to 10% of the data obtained by means of the successive radar measurements (CIR).

8. Method according to one of the above claims, wherein determining a threshold value (SW*) for the saliency value (SW), according to which the selected point (Tap) is evaluated as salient (true) or not salient (false).

9. The method according to one of the preceding claims, wherein, in which, when the selected point (Tap) is evaluated as salient (true) or not salient (false) for a desired vehicle function inside and / or outside the vehicle (F), the position of the potential target (Z) relative to the vehicle (F) is taken into account, which can be inside or outside the vehicle (F), and / or in which, for a vehicle function inside the vehicle (F), potential targets outside the vehicle (F) can be ignored.

10. The method according to one of the preceding claims, wherein the method is implemented for tracking objects, such as vehicles and / or persons, outside and / or inside the vehicle (F), and / or in which, for a vehicle function outside the vehicle (F), potential targets inside the vehicle (F) can be ignored, and / or in which, when an incorrect behavior of an object is recognized in tracking an object outside the vehicle (F), measures are introduced, such as activating an external camera, switching on lighting, outputting an alarm, etc. and / or in which, when tracking an object from outside the vehicle (F) into an interior space of the vehicle, an identity verification is triggered.

11. The method according to one of the preceding claims, wherein the method is implemented for operating a plurality of sensor devices (100), in which, in particular, the information of the plurality of sensor devices (100) is aggregated in order to provide different vehicle functions outside and / or inside the vehicle (F).

12. Computer program product, comprising instructions which, when the computer program product is executed by a computer, cause the computer to implement the method according to one of the preceding claims.

13. A control unit (ECU) having a storage unit and a calculation unit, in which a code is stored in the storage unit, wherein the method according to one of the preceding claims 1 to 11 is implemented when the code is executed by the computing unit, in which, in particular, when the method is implemented for operating a plurality of sensor devices (100), a control unit (ECU) of one sensor device (100) of the plurality of sensor devices (100) has a master function or a plurality of control units (ECU) each have an evaluation function in order to aggregate the information of the plurality of sensor devices (100).

14. Sensor device (100), in particular UWB sensor device, having a control unit (ECU) according to the preceding claim.

Images