Method and device for ascertaining tangential speed of object

Abstract

The invention relates to a device for determining speed information relating to the tangential speed of an object in the environment of a radar sensor. The radar sensor is designed to transmit a transmission signal and to receive a respective plurality of reception signals at a plurality of reception antennas, each of which depends on the transmission signal. The transmission signal has a frequency ramp in each of N temporally consecutive time intervals. The device is configured to determine angle information relating to a detection angle of the object for each of the N time intervals on the basis of the plurality of reception signals and on the basis of the transmission signals. The device is further configured to determine speed information relating to a tangential speed of the object on the basis of a time series of angle information relating to the detected angle.

Description

Field of the Invention

[0001] The present invention relates to a method and a corresponding device for determining the tangential velocity of an object based on sensor data of a radar sensor. Background Art

[0002] At least partially automated vehicles have one or more environmental sensors which are arranged to detect sensor data relating to the vehicle's surroundings. The sensor data can be evaluated to provide a driving function by means of which the vehicle can be longitudinally and / or laterally guided at least partially automatically. In particular, the vehicle can include a radar sensor as an environmental sensor to detect objects in the vehicle's environment. Here, based on the sensor data of the radar sensor, in particular based on the Doppler effect, the radial velocity of each object (in the radial direction with respect to the radar sensor) can also be determined. Summary of the Invention

[0003] This document relates to the following technical task of determining reliable and accurate information relating to the tangential velocity (tangential to the detection direction of the radar sensor) of an object based on the sensor data of the radar sensor.

[0004] This task is solved by each independent claim. Advantageous embodiments are described in particular in the dependent claims. It should be noted that the additional features of the claims dependent on an independent claim may not include the features of the independent claim or may only be combined with some of the features of the independent claim, and constitute a separate invention independent of all combinations of the features of the independent claim, which may be the subject of an independent claim, a divisional application or a later application. This also applies to the technical teachings described in the specification, which may constitute an invention independent of the features of the independent patent claims.

[0005] According to one aspect, a device for determining velocity information relating to the tangential velocity of an object in the environment of a radar sensor is described. The radar sensor can be arranged in a vehicle. The tangential velocity of the object can be the velocity of the object's movement tangential to the detection direction of the radar sensor (from the detection range of the radar sensor).

[0006] The radar sensor can be configured to transmit a transmission signal and receive a corresponding plurality of reception signals at a plurality of reception antennas, the plurality of reception signals each depending on the transmission signal (due to the reflection of the transmission signal at the object). The reception antennas can be arranged offset from each other transversely and / or laterally (with respect to the detection range and / or detection direction of the radar sensor). Here, the reception antennas can be equidistant from each other.

[0007] The transmitted signal can have frequency ramps in N successive time intervals respectively (N>1, for example N = 10 or more, or N = 20 or more, or N = 50 or more). In particular, the transmitted signal can consist of a time series of frequency ramps. Here, the frequency ramps can have respectively multiple different frequencies of the transmitted signal. In other words, the frequency of the transmitted signal can vary along the frequency ramp, for example starting from a first frequency (which may be smooth or stepwise) to a second (higher) frequency.

[0008] The device is arranged to determine, for each of the N time intervals, angle information related to the detection angle of the object in the corresponding time interval based on the plurality of received signals and based on the transmitted signal (for example by mixing). In other words, a (time) series of angle information related to the detection angle of the object in N successive time intervals can be determined. For this purpose, a digital beamforming method can be used.

[0009] The device is further arranged to determine velocity information related to the tangential velocity of the object based on the time series of angle information related to the detection angle. The velocity information of the object related to the tangential velocity, in particular the tangential velocity, can be determined along the N time intervals especially based on the change of the angle information, in particular based on the change of the detection angle.

[0010] The device can be arranged, for example, to determine the (time) gradient, and the angle information, in particular the (average) detection angle, varies (for example, increases or decreases) along the N time intervals with this gradient. The velocity information related to the tangential velocity of the object can be determined in a particularly precise manner based on the gradient.

[0011] The present invention describes a device which is arranged to determine the time variation of the detection angle of an object based on a plurality of received signals. Based on this time variation, the tangential velocity of the object can be determined in an effective and precise manner.

[0012] The device can be arranged to determine Q intermediate frequency signals based on the plurality of received signals and based on the transmitted signal by mixing the received signals with the transmitted signal respectively, where Q>1 (for example Q = 10 or more, or Q = 20 or more, or Q = 50 or more). The time series of angle information related to the detection angle, in particular the time series of the detection angle, can be determined in a precise manner based on the Q intermediate frequency signals.

[0013] The device can be set to a determined time interval for N time intervals, and for each of the Q intermediate frequency signals, respectively obtain the phase value of the corresponding intermediate frequency signal in the determined time interval. Therefore, for the Q intermediate frequency signals, the phase values of each intermediate frequency signal in the determined time interval can be respectively obtained, so as to obtain a sequence of Q phase values (the corresponding sequence of Q combined index values along the combined dimension, where the combined index identifies different combinations of the corresponding received signal and transmitted signal; in addition, the combined index can display or indicate different lateral distances between different receiving antennas).

[0014] Based on the Q values of the phases of the Q intermediate frequency signals in the determined time interval, angle information related to the detection angle of the object in the determined time interval can be obtained. In particular, the device can be set to obtain a spectrum based on the band transformation of the sequence of the Q values of the phases of the Q intermediate frequency signals in the determined time interval. The angle information related to the detection angle of the object in the determined time interval can be accurately obtained based on the obtained spectrum (for example, the frequency corresponding to the maximum value of the spectrum).

[0015] The device can be set to obtain a (possibly three-dimensional) radar tensor with multiple units based on multiple received signals and based on the transmitted signal. Based on multiple received signals and based on the transmitted signal, corresponding multiple inter-band signals can be obtained (in particular, by mixing each received signal with the transmitted signal). The measured values of the multiple units of the radar tensor can be obtained based on multiple intermediate frequency signals, in particular by sampling the multiple intermediate frequency signals.

[0016] The radar tensor can have a long-term dimension. The radar tensor can respectively include (for example, two-dimensional) sub-tensors along the long-term dimension for each of the N time intervals.

[0017] The sub-tensor can have a short-term dimension, where the sub-tensor respectively includes units along the short-term dimension for K sampling time points of the frequency ramp (where K>1, for example, K = 20 or greater, or K = 50 or greater, or K = 200 or greater).

[0018] In addition, the sub-tensor can have a combined dimension, where the sub-tensor respectively includes units along the combined dimension for each of the Q different combinations of the received signal and the transmitted signal.

[0019] The device can be set to respectively perform a band transformation, in particular a Fourier transform, along the short-term dimension for each of the N sub-tensors in the corresponding N time intervals, so as to obtain distance information related to the radial distance of the object (along the detection direction of the radar sensor). The spectrum along the short-term dimension can be obtained through the band transformation. The radial distance of the object can be realized by the frequency at which the spectrum has a maximum value.

[0020] The device can also be set to perform a band transformation, in particular a Fourier transformation, along the combined dimension, in particular (possibly only) for the cells corresponding to the radial distance of the object to be determined, in order to determine angle information related to the detection angle of the object for the corresponding time interval. By means of the band transformation, the frequency spectrum can be determined along the combined dimension. The angle information, in particular the detection angle, can be indicated by the frequency at which the frequency spectrum has a maximum value.

[0021] Therefore, a band transformation is performed along the combined dimension (after the band transformation along the short-term dimension and before the band transformation along the long-term dimension) in order to determine, for each time interval, angle information related to the detection angle of the object in the corresponding time interval. Thus, the tangential velocity of the object can be determined in a particularly precise manner.

[0022] The device can be set to perform a band transformation, in particular a Fourier transformation, along the long-term dimension, in particular (possibly only) for the cells corresponding to the radial distance of the object to be determined, without having previously performed a band transformation along the combined dimension. Velocity information related to the radial velocity of the object can be determined based on the band transformation along the long-term dimension.

[0023] The device can be set to, based on L respective time series for determining angle information related to the detection angle of the object, which are obtained from L radar sensors arranged adjacent to one another (in terms of the detection range or in terms of the detection direction), where L > 1 (for example, L = 2 or more, or L = 3 or more). Velocity information related to the tangential velocity of the object can be determined in a particularly precise manner based on the L time series of angle information related to the detection angle of the object.

[0024] As described above, the radar sensors can be part of a vehicle. The device can be set to provide a driving function for the automatic longitudinal and / or lateral guidance of the vehicle based on the velocity information.

[0025] According to another aspect, a (road) motor vehicle (in particular a passenger car or a lorry or a bus or a motorcycle) is described, which comprises the device described herein.

[0026] According to another aspect, a method for determining velocity information related to the tangential velocity of an object in the environment of a radar sensor (where the radar sensor can be arranged in a vehicle) is described. The radar sensor is designed to transmit a transmission signal and to receive, at a plurality of receiving antennas, a respective plurality of received signals, which depend on the transmission signal (due to reflections at the object). The transmission signal has a frequency ramp in each of N time intervals that are successive in time.

[0027] The method includes obtaining angle information related to the detection angle of an object for N time intervals respectively based on a plurality of received signals and based on a transmitted signal. The method further includes obtaining velocity information related to the tangential velocity of the object based on a time series of the angle information related to the detection angle.

[0028] According to another aspect, a software (SW) program is described. The SW program can be configured to be executed on a processor (e.g., on a controller of a vehicle), and thereby execute the methods described in this document.

[0029] According to another aspect, a storage medium is described. The storage medium can include an SW program that is configured to be executed on a processor, and thereby execute the methods described herein.

[0030] It should be noted that the methods, devices, and systems described herein can be used not only alone but also in combination with other methods, devices, and systems described herein. In addition, any aspects of the methods, devices, and systems described herein can be combined with each other in various ways. In particular, the features of the claims can be combined in various ways. In addition, the features listed in parentheses should be understood as optional features. Description of the Drawings

[0031] Hereinafter, the present invention will be described in more detail based on embodiments. The drawings show:

[0032] Figure 1a An exemplary vehicle having a radar sensor is shown;

[0033] Figure 1b An exemplary transmitted signal of the radar sensor is shown;

[0034] Figure 2a An exemplary measurement value tensor of the radar sensor is shown;

[0035] Figure 2b An exemplary frequency domain tensor of the measurement value tensor of the radar sensor is shown;

[0036] Figure 3 An exemplary modified frequency domain tensor is shown;

[0037] Figure 4 A flowchart of an exemplary method for obtaining velocity information related to the tangential velocity of an object is shown; and

[0038] Figure 5 An example distribution of signal energy for different ramp indices and different antenna combination indices is shown. Detailed Description of the Embodiments

[0039] As described at the beginning, this document relates to precisely determining the tangential velocity of an object arranged in a (motor) vehicle environment. In this context, Figure 1a An exemplary vehicle 100 is shown having at least one radar sensor 102, which is arranged to detect sensor data related to the environment of the vehicle 100, in particular sensor data related to the environment in front of the vehicle 100. The radar sensor 102 can be arranged to transmit a transmission signal and receive a reception signal depending on the transmission signal. The reception signal can be based on the reflection of the transmission signal on an object 105 in the environment of the vehicle 100.

[0040] The radar sensor 102 can have a defined detection range 120, where the detection range 120 has a plurality of different detection directions 121. The plurality of different detection directions 121 can be defined in the polar coordinates of a polar coordinate system, where the radar sensor 102 is arranged at the origin of the polar coordinate system. The different detection directions 121 can have azimuth angles within a defined azimuth angle range. In addition, the different detection directions 121 can have elevation angles within a defined elevation angle range. Thus, the detection range 120 of the radar sensor 102 can be defined by a combination of the azimuth angle range and the elevation angle range. In addition, there may be a limitation of the detection range in terms of the maximum possible radial distance of detection. In addition, the radar sensor 102 can have a defined angular resolution related to the possible azimuth angles and / or to the possible elevation angles.

[0041] In Figure 1a an exemplary reception direction 121 with a defined elevation angle 122 is shown. In general, the reception direction 121 can have a defined azimuth angle and / or elevation angle 122.

[0042] The radar sensor 102 can be arranged to detect a detection point 111 in a defined detection direction 121 (i.e., for a defined azimuth angle 122 and a defined elevation angle 122) based on the transmission signal when the transmission signal is reflected by the object 105 or the ground 110 on which the vehicle 100 is traveling. On the other hand, if the transmitted signal is not reflected, the radar sensor 102 generally does not detect the detection point 111. Thus, the radar sensor 102 can provide a point cloud of detection points 111 for the detection range 120 of the radar sensor 102, through which one or more objects 105 and possibly the ground 110 in the environment of the vehicle 100 can be displayed.

[0043] The (control) device 101 of the vehicle 100 can be set to evaluate the sensor data of the radar sensor 102 (in particular the detection point cloud 111), for example, to create an environmental model of the environment of the vehicle 100. The device 101 can in particular be set to provide one or more driving functions based on the sensor data of the radar sensor 102, by means of which the vehicle 100 can be longitudinally and / or laterally guided at least partially or fully automatically. For this purpose, one or more longitudinal and / or lateral guidance actuators 103 of the vehicle 100 (such as drive motors, braking devices and / or steering devices) can be actuated.

[0044] Figure 1b An exemplary transmission signal 130 that can be transmitted by the transmitting antenna of the radar sensor 102 is shown. The transmitting antenna can have a defined radiation characteristic by means of which the detection range 120 of the radar sensor 102 is defined and / or specified. The transmission signal 130 can have a sequence of frequency ramps 135, in which within the frequency ramp 135, the frequency 132 of the transmission signal 130 initially changes continuously or step by step (starting from a first frequency), for example increases (reaching a second frequency), and suddenly returns to the first frequency at the end of the frequency ramp 135.

[0045] The transmission signal 130 can have N frequency ramps 135 along the time 131, for example N = 10 or more, or N = 100 or more. The frequency ramp 135 can have a defined ramp duration T here. Each frequency ramp 135 can be identified by a ramp index n 133, where n = 1,..., N. The long-term dimension can be defined by the ramp index n 133.

[0046] Each frequency ramp 135 can be divided into K sampling times, for example K = 10 or greater, or K = 100 or greater. Thus, the individual sampling times are spaced apart from each other by T / K in time. Each sampling time point within the frequency ramp 135 can be represented by a sampling index k 134, where k = 1,..., K. The short-term dimension can be defined by the sampling index k 134.

[0047] The radar sensor 102 can also include a set of receiving antennas (for example, 3 or more, or 5 or more, or 10 or more receiving antennas), by means of which the received signal caused by the reflection of the transmission signal 130 can be received. The receiving antennas can be arranged spatially offset from each other, which makes it possible to determine the incident direction of the received signal on the set of receiving antennas by means of the reception time offset and / or the phase of the received signal. The input direction of the received signal can be regarded here as the detection direction 121 of the detection point 111 associated with the received signal.

[0048] The radar sensor 102 may have a set of transmitting antennas, which are arranged to transmit different transmission signals 130 from each other when necessary. The transmitting antennas may be arranged offset from each other spatially.

[0049] By using multiple receiving antennas and / or multiple transmitting antennas, different incident directions and thus different detection directions 121 can be distinguished, which enables the detection point 111 to be detected in different detection directions 121, i.e., at different detection angles 122. The radar sensor 102 may be configured to distinguish Q different detection directions 121 or detection angles 122 (e.g., Q = 10 or more, or Q = 100 or more). The Q detection angles 122 may be based on a combination of L transmitting antennas and M receiving antennas (e.g., Q = L x M).

[0050] Therefore, Q received signals of Q different antenna combinations can be detected by the radar sensor 102. Each received signal is respectively based on a time series of a frequency ramp 135 (also referred to as a chirp). Each received signal may be sampled at a sampling frequency f (e.g., f = K / T) (possibly immediately after mixing with the original transmission signal 130 to generate a corresponding intermediate frequency signal for further signal processing) to provide a time series of measurement values. The time series of measurement values of the received signal (or intermediate frequency signal) may be divided into N groups, each group having K measurement values. Therefore, the radar sensor 102 can provide Q different sequences of measurement values (for Q different antenna combinations or for Q different detection angles), where each sequence of measurement values includes N temporally successive groups, each group having K measurement values.

[0051] The K x N x Q measurement values can be combined (e.g., for illustration purposes) in a measurement value or radar tensor 200, as shown exemplarily in Figure 2a Here, the measurement value and / or the radar tensor 200 may have,

[0052] · a (long-term) dimension of the ramp index n 133;

[0053] · a (short-term) dimension of the sampling index k 134; and / or

[0054] · an (antenna combination) dimension of the combination index q 222 (for the corresponding antenna combination), where q = 1,..., Q.

[0055] Each cell 201 of the measurement value tensor 200 may respectively have a measurement value x(k, n, q).

[0056] It can be shown that the frequency of the intermediate frequency signal (obtained by mixing the transmitted signal and the received signal) depends on the radial distance r of the object 105 and the radial velocity v of the object 105 that generates the received signal (by reflection of the transmitted signal). Furthermore, it can be shown that the phase of the intermediate frequency signal is proportional to the combined index q 222 (when using equidistant (virtual) antennas), where the proportionality factor depends on the angle of incidence of the received signal and thus on the detection angle 122 of the object 105.

[0057] To evaluate the measured value x(k, n, q) of the intermediate frequency signal detected by the radar sensor 102, one or more time-domain frequency-domain transforms, in particular the fast Fourier transform (FFT), can be performed. In particular, in a first step, the transform can be performed along the sampling index or short-term dimension, which makes it possible to determine the radial distance r of the detection point 111 or the object 105.

[0058] The transform along the sampling index or short-term dimension provides a tensor of partial transforms, which has elements x'(r,n,q), where r corresponds to the radial distance. Relatively high values of x'(r,n,q) indicate that the detection point 111 is located at the time point indicated by n, for the antenna combination indicated by q, and at the distance indicated by r. Relatively low values of x'(r,n,q) indicate that there is no detection point 111 at the time point indicated by n, for the antenna combination indicated by q, and at the distance indicated by r.

[0059] After the transform along the sampling index or short-term dimension, in a second step, the transform can be performed along the ramp index or long-term dimension. The transform along the ramp index or long-term dimension can here be limited to a sub-tensor (in particular a matrix) of the partially transformed tensor, where there are relatively high values of x'(r,n,q) (for example, values of x'(r,n,q) exceeding a determined threshold).

[0060] By means of this transform along the ramp index or long-term dimension, the velocity v of the change in the radial distance r of the detection point 111 or the object 105 can be determined, that is, the radial velocity v of the detection point 111 can be determined.

[0061] The additional transform performed along the ramp index or long-term dimension provides a (sub-)tensor of partial transforms with elements x"(r,v,q), where v corresponds to the radial velocity. Relatively high values of x"(r,v,q) indicate that there is a detection point 111 for the antenna combination indicated by q and at the distance indicated by r, which is moving with a radial velocity v. Relatively low values of x"(r,v,q) indicate that there is no detection point 111 for the antenna combination indicated by q and at the distance indicated by r.

[0062] By further transformation along the combined index dimension, the detection angle 122 of the detection point 111 and / or the object 105 can be identified. In particular, a transformed tensor 210 with elements x'''(r, v, s) 211 can be provided (see Figure 2b ), where s corresponds to the angle index 232 for identifying the corresponding detection angle 122. As described above, the phase of the intermediate frequency signal is proportional to the combined index q222, by which each virtual antenna (i.e., each antenna combination) can be identified. Therefore, the scaling factor depending on the detection angle 122 corresponds to the "frequency" relative to the combined index q222. This frequency and the detection angle 122 based on this frequency can be obtained along the combined index dimension according to the frequency band transformation. Relatively high values of x'''(r, v, s) indicate that there is a detection point 111 at the detection angle 122 indicated by s and at the distance indicated by r, moving with a radial velocity v. Relatively low values of x'''(r, v, s) indicate that there is no detection point 111 at the detection angle 122 indicated by q and at the distance indicated by r.

[0063] In Figure 2b , the distance index r is represented by the reference numeral 234, the velocity index v is represented by the reference numeral 233, and the angle index s is represented by the reference numeral 232.

[0064] To obtain the tangential velocity of the detection point 111 or the object 105 corresponding to the detection point 111, the partially transformed tensor with elements x'(r, n, q) can be transformed along the combined dimension by a transformation along the sampling index or the short-term dimension, in order to obtain a tensor with elements x'*(r, n, s). Here, the transformation along the antenna combination dimension can be limited to a sub-tensor (in particular, a matrix) of the partially transformed tensor for which there are relatively high values of x'(r, n, q) (for example, values of x'(r, n, q) exceeding a determined threshold).

[0065] Figure 5 An exemplary frequency band transformation along the combined dimension is shown in Figure 5The long-term dimension with the ramp index 133 extends in the horizontal direction, and the combined dimension with the combined index 222 extends in the vertical direction. Each spectrum 500 is obtained by a band transformation along the sampling and / or short-term dimension for a corresponding determined ramp index 133 and for a corresponding determined combined index 222. Here, the spectrum 500 shows the amplitude or signal energy 502 as a function of the frequency 501, where different frequencies 501 respectively correspond to different radial distances r. Thus, the frequency 501 can correspond to the distance index 234 of the partial transformation tensor x'(r,n,q). The frequency 505 corresponds to the amplitude and / or energy maximum 505 of the spectrum 500, indicating the radial distance r of the object 105 or the detection point 111.

[0066] In Figure 5 an exemplary spectrum 500 for the object 105 is shown, which has a fixed radial distance r and moves with a determined tangential velocity. The tangential movement of the object 105 causes the detection angle 122 of the object 105 to change along the ramp index or the long-term dimension. As described above, the phase of the intermediate frequency signal is proportional to the combined index 222, where the proportionality factor depends on the detection angle 122.

[0067] Figure 5 Example values of the phase of each intermediate frequency signal are shown. Here, the intermediate frequency signals in the left column (for the ramp index n = l) have a phase value of 0° for all combined indices q = l,...,Q, which indicates that the detection angle 122 of the object 105 is 0°. Due to the tangential movement of the object 105, the detection angle 122 of the object 105 changes as the ramp index increases, which causes the intermediate frequency signals in the columns n > l to have phase values > 0° respectively. Here, the phase values of the intermediate frequency signals within a column (i.e., for a determined ramp index n) increase as the combined index q increases. The increase in the phase values within a column depends here on the incident angle 122 of the object 105 at the time point corresponding to the determined ramp index n.

[0068] By a band transformation along the combined dimension, the frequency 511 can be obtained separately for each ramp index n, which shows the incident angle 122 of the object 105 at the time point corresponding to each ramp index n. The frequency 511 is generally proportional to the sine or cosine of the incident angle 122.

[0069] Thus, a time series of the frequency 511 can be obtained, and based on this, a time series of the incident angle 122 can be obtained (for the corresponding ramp index sequence n = 1,...,N). Based on the time series of the incident angle 122, the tangential velocity value of the object 105 can be obtained (e.g., by linear regression and / or band transformation).

[0070] Thus, a tensor with cells x'(r,n,q) that is partially transformed by transformation along the sampling index or short-term dimension can be transformed (e.g., for a part of the tensor) along the combined dimension to obtain a (sub)tensor with cells x'*(r,n,s). Relatively high values of x'*(r,n,s) indicate that there is a detection point 111 at the time point indicated by n, for the detection angle 122 indicated by s, and at the distance indicated by r. Relatively low values of x'*(r,n,s) indicate that there is no detection point 111 at the time point indicated by n, for the detection angle 122 indicated by s, and at the distance indicated by r.

[0071] This tensor can be transformed (e.g., partially) along the ramp index or long-term dimension to provide a fully transformed tensor 310 with cells x'**(r,w,s) 311 (see Figure 3 ), where w indicates the tangential velocity of the detection point 111. In particular, w indicates the speed at which the detection angle 122 of the detection point 111 changes. This (angular) velocity provides information about the velocity of the detection point 111 in the tangential direction (relative to the detection direction 121 of the detection point 111).

[0072] In Figure 3 , the distance index r is represented by reference numeral 234, the velocity index w (for tangential velocity) is represented by reference numeral 333, and the angle index s is represented by reference numeral 232.

[0073] As described above, the radar sensor 102 can directly obtain at most four dimensions by measurement, in particular radial distance, azimuth angle, elevation angle, and / or radial velocity. These dimensions can be transformed into a Cartesian coordinate system if needed. The object 105 to be measured moves tangentially towards the vehicle 100 and thus towards the radar sensor 102 (e.g., a vehicle in front that is moving laterally due to a lane change). The tangential movement of this object 105 can usually only be detected by observing the object 105 over time (i.e., by tracking), rather than by direct measurement. As a result, the determination of the tangential velocity of the object 105 is delayed and averaged, which can lead to disadvantages in the recognition of the surrounding environment and thus to a deterioration of the functional characteristics of the driving function. For example, the speed of the driving maneuvers of the vehicle 100 can be restricted thereby. In addition, the accuracy of determining the tangential velocity of the object 105 may be limited, which may lead to restrictions on the driving maneuvers of the vehicle 100.

[0074] As described in this document, the direct determination of the tangential velocity can be achieved by analyzing the angle spectrum of each frequency ramp (i.e., chirp) in a time series of frequency ramps. This can be similar to the Doppler determination performed by a radar sensor under fast chirp FMCW (Frequency Modulated Continuous Wave), where the intermediate frequency phase from one chirp to another corresponds to the Doppler frequency.

[0075] In an angle formation method, for example, to determine the azimuth angle, digital beamforming can be used to generate a power spectrum that has a peak at the most likely angle 122 of the detection point 111. This is typically done after calculating the two-dimensional FFT of all FMCW chirps to determine the range and Doppler velocity of one or more objects 105.

[0076] As described herein, it is already possible to observe the (phase) differences in the beamforming power spectrum based on the individual chirps on all (virtual) antennas 222. The time offset of the chirps causes the phase relationship between the antennas 222 to change over time 131. Thus, an object 105 moving tangentially generates a (spatial) offset spectrum over time 131. The period considered is usually small, making it unlikely for the reflection point to migrate because the geometric situation only changes slightly. Therefore, the temporal changes in the beamforming spectrum can be accurately assigned to an object 105.

[0077] Depending on the characteristics, time tracking of the changes in the beamforming power spectrum can be achieved by comparison (e.g., by tracking), or in cases where the differences are relatively small, by frequency transformation (e.g., FFT).

[0078] Combined Figure 3 , the implementation of the FFT is described. After the first FFT (for obtaining the radial distance), an FFT is performed on the (virtual) antenna channels q based on the individual chirps or frequency ramps to create an angle spectrum. Along the chirps (i.e., along the ramp index n), the change in the angle spectrum can be obtained by means of another FFT. The result represents the tangential velocity obtained by using the FFT for each range / angle peak. In particular, the frequency corresponding to the tangential velocity of the detection point 111 previously identified in the range or distance spectrum is obtained.

[0079] The above signal processing steps can be performed in parallel or serially if necessary to obtain the radial velocity and the tangential velocity (Doppler FFT along the chirps, angle FFT along the antenna channels based on all chirps).

[0080] If necessary, multiple (laterally distributed) radar sensors 102 can be used, which operate coherently or incoherently. Thus, the tangential velocity of the object 105 can be obtained with increased accuracy.

[0081] Figure 4 A flowchart of a (possibly computer-implemented) method 400 for obtaining tangential velocity information related to an object 105 in the environment of a radar sensor 102 is shown. The tangential velocity of the object 105 can be the velocity of the object 105 moving tangentially along the detection direction 121 of the radar sensor 102.

[0082] The radar sensor 102 is configured to transmit a transmission signal 130 and receive a corresponding plurality of received signals at a plurality of receiving antennas, where the received signals respectively depend on the transmission signal 130 (e.g., due to reflections at the object 105). The transmission signal 130 has a frequency ramp 135 in each of N temporally successive time intervals. The frequency ramp 135 can here have a frequency that varies within the corresponding time interval (e.g., such that a transmission signal 130 with a sawtooth frequency modulation is generated).

[0083] The method 400 includes respectively determining 401 angle information 511 (within the corresponding time interval) related to the detection angle 122 for the object 105 based on the plurality of received signals and based on the transmission signal 130 (e.g., by mixing the plurality of received signals with the transmission signal 130). The angle information 511 can for example correspond to a frequency indicating the detection angle 122. For example, the frequency determined (according to a frequency band transformation, such as a Fourier transform) can be proportional to the cosine or sine of the detection angle 122.

[0084] The method 400 further includes determining 402 velocity information related to the tangential velocity of the object 105 based on the time series of the angle information 511 with respect to the detection angle 122. Here, in particular, the gradient of the change of the angle information 511, especially the detection angle 122, can be determined to determine the tangential velocity of the object 105. For example, the angular velocity of the change of the detection angle 122 over time (i.e., along the N time intervals) can be determined. Based on the angular velocity (taking into account the radial distance of the object 105), the tangential velocity of the object 105 can be determined in an effective and accurate manner.

[0085] The invention is not limited to the illustrated embodiments. In particular, it should be noted that the description and the drawings only exemplarily illustrate the principles of the proposed methods, devices, and systems.

Claims

1. An apparatus (101) for obtaining velocity information related to the tangential velocity of an object (105) in the environment of a radar sensor (102); wherein the radar sensor (102) is configured to transmit a transmission signal (130) and receive a corresponding plurality of received signals at a plurality of receiving antennas, the plurality of received signals respectively depending on the transmission signal (130); wherein the transmission signal (130) has a frequency ramp (135) in each of N time intervals that are successive in time; wherein the apparatus (101) is arranged to - obtain angle information (511) related to the detection angle (122) for the object (105) for each of the N time intervals, based on the plurality of received signals and based on the transmission signal (130); and - obtain velocity information related to the tangential velocity of the object (105) based on a time series of the angle information (511) related to the detection angle (122).

2. The apparatus (101) according to claim 1, wherein the apparatus (101) is arranged to obtain the velocity information related to the tangential velocity of the object (105) along the N time intervals based on a change in the angle information (511), in particular based on a change in the detection angle (122).

3. The apparatus (101) according to claim 2, wherein the apparatus (101) is arranged to - obtain a gradient, with which the angle information (511), in particular the detection angle (122), changes along the N time intervals; and - obtain the velocity information related to the tangential velocity of the object (105) based on the gradient.

4. The apparatus (101) according to any one of the preceding claims, wherein the apparatus (101) is arranged to - obtain Q intermediate frequency signals by mixing the received signals with the transmission signal (130) respectively, based on the plurality of received signals and based on the transmission signal (130), where Q > 1; and - obtain the time series of the angle information (511) related to the detection angle (122), in particular the time series of the detection angle (122), based on the Q intermediate frequency signals.

5. The apparatus (101) according to claim 4, wherein the apparatus (101) is arranged such that, for a determined time interval among the N time intervals, - obtain, for each of the Q intermediate frequency signals, the value of the phase of the corresponding intermediate frequency signal in the determined time interval; and - obtain the angle information (511) related to the detection angle (122) for the object (105) in the determined time interval based on the Q values of the phases of the Q intermediate frequency signals in the determined time interval.

6. The apparatus (101) according to claim 5, wherein the apparatus (101) is arranged to - obtain a frequency spectrum based on a band transformation of a sequence of the Q values of the phases of the Q intermediate frequency signals in the determined time interval; and - Based on the obtained spectrum, obtain the angle information (511) related to the detection angle (122) of the object (105) in the determined time interval.

7. The apparatus (101) according to any one of the preceding claims, wherein - the apparatus (101) is configured to obtain a radar tensor (200) having a plurality of cells (201) based on the plurality of received signals and based on the transmitted signal (130); - the radar tensor (200) has a long-term dimension; - the radar tensor (200) includes sub-tensors having cells (201) respectively for each of the N time intervals along the long-term dimension; - the sub-tensor has a short-term dimension; - the sub-tensor includes cells (201) respectively for K sampling time points (123) of one frequency ramp (135) along the short-term dimension; - the sub-tensor has a combined dimension; - the sub-tensor includes cells (201) respectively for each of Q different combinations of the received signal and the transmitted signal along the combined dimension; and - the apparatus (101) is configured to, for each of the N sub-tensors, for the corresponding N time intervals, - perform a band transformation, in particular a Fourier transform, respectively along the short-term dimension to obtain distance information related to the radial distance of the object (105); and - in particular for the cells (201) corresponding to the obtained radial distance of the object (105), perform a band transformation, in particular a Fourier transform, along the combined dimension to obtain the angle information (511) related to the detection angle (122) of the object (105) for the corresponding time interval.

8. The apparatus (101) according to claim 7, wherein the apparatus (101) is configured to - in particular for the cells (201) corresponding to the obtained radial distance of the object (105), perform a band transformation, in particular a Fourier transform, along the long-term dimension without previously performing a band transformation along the combined dimension; and - based on the band transformation along the long-term dimension, obtain velocity information about the radial velocity of the object (105).

9. The apparatus (101) according to any one of claims 7 to 8, wherein the apparatus (101) is configured to - obtain corresponding multiple intermediate frequency signals based on the plurality of received signals and based on the transmitted signal (130), in particular by using a mixer; and - based on the multiple intermediate frequency signals, in particular by sampling the multiple intermediate frequency signals, obtain measurement values of the plurality of cells (201) of the radar tensor (200).

10. The apparatus (101) according to any one of the preceding claims, wherein the tangential velocity of the object (105) is the velocity of the movement of the object (105) tangent to the detection direction (121) of the radar sensor (102).

11. The device (101) according to any one of the preceding claims, wherein the device (101) is configured to - obtain L respective time series of angle information (511) related to the detection angle (122) of the object (105) based on L radar sensors (102) arranged adjacent to each other, where L > 1; and - obtain velocity information related to the tangential velocity of the object (105) based on the L time series of angle information (511) related to the detection angle (122) of the object (105).

12. A method (400) for obtaining velocity information related to the tangential velocity of an object (105) in the environment of a radar sensor (102); wherein the radar sensor (102) is configured to transmit a transmission signal (130) and receive a respective plurality of received signals at a plurality of receiving antennas, the plurality of received signals respectively depending on the transmission signal (130); wherein the transmission signal (130) has a frequency ramp (135) in N time intervals that are successive in time; wherein the method (400) comprises: - obtaining (401) angle information (511) related to the detection angle (122) of the object (105) for each of the N time intervals respectively based on the plurality of received signals and based on the transmission signal (130); and - obtaining (402) velocity information related to the tangential velocity of the object (105) based on a time series of angle information (511) related to the detection angle (122).

Images