METHOD AND DEVICE FOR DETECTING CRITICAL TRANSVERSE MOVEMENTS

Abstract

A method for detecting critical transverse movements. The method includes the following steps: emitting a CW radar signal and generating radar data based on the received reflected CW radar signal with the aid of a radar device; ascertaining collision-relevant spectral ranges of the radar data as a function of an ego velocity of the radar device; ascertaining a time dependency of a relative velocity and of an object angle of an object by evaluating the radar data in the ascertained spectral ranges; and detecting a critical transverse movement of the object using the time dependency of the relative velocity and of the object angle of the object.

Claims

1-10 (canceled)

11. A method for detecting critical transverse movements, comprising the following steps: emitting a continuous wave (CW) radar signal and generating radar data based on a received reflected CW radar signal using a radar device; ascertaining collision-relevant spectral ranges of the radar data as a function of an ego velocity of the radar device; ascertaining a time dependency of a relative velocity of an object and of an object angle of the object, by evaluating the radar data in the ascertained spectral ranges; and detecting a critical transverse movement of the object using the time dependency of the relative velocity of the object and of the object angle of the object.

12. The method as recited in claim 11, wherein the radar device includes a plurality of antenna devices having detection ranges differing from one another, an activation of the antenna devices being a function of whether the critical transverse movement of the object is detected.

13. The method as recited in claim 12, wherein in the event the critical transverse movement of the object is detected, at least one of the antenna devices is activated to emit a frequency modulated continuous wave (FMCW) radar signal, and a movement of the object being more precisely determined based on a received reflected FMCW radar signal.

14. The method as recited in claim 11, wherein the radar device includes a plurality of receiver units, and the object angle is calculated based on phase differences of the reflected CW radar signal received by various ones of the receiver units.

15. The method as recited in claim 11, wherein the collision-relevant spectral ranges are determined in such a way that only objects having a negative relative velocity are considered.

16. The method as recited in claim 11, wherein the collision-relevant spectral ranges are determined under an assumption that the ego velocity of the radar device, the relative velocity of the object, and the object angle of the object, are constant.

17. The method as recited in claim 11, wherein collision-critical relative velocities and corresponding collision-critical object angles are determined as a function of the ego velocity of the radar device, of an intersecting angle between a trajectory of the object and a trajectory of the radar device, and of an object velocity of the object, and the collision-relevant spectral ranges are limited to those of the collision-critical ego velocities and collision-critical object angles, which correspond to the ego velocity of the radar device, to a predefined range of the intersecting angles, and to a predefined range of the object velocity.

18. The method as recited in claim 17, wherein the collision-critical relative velocities and collision-critical object angles are further determined taking finite expansions of the object and of a vehicle that includes the radar device, into consideration.

19. The method as recited in claim 11, wherein the critical transverse movement of the object is detected in an event relative changes of the relative velocity and of the object angle do not exceed predefined threshold values.

20. A device for detecting critical transverse movements, comprising: a radar device configured to emit a continuous wave (CW) radar signal and to generate radar data based on a received reflected CW radar signal; and a computing device configured to ascertain collision-relevant spectral ranges of the radar data as a function of an ego velocity of the radar device, to ascertain a time dependency of a relative velocity of an object and of an object angle of the object, by evaluating the radar data in the ascertained spectral ranges, and to detect a critical transverse movement of the object using the time dependency of the relative velocity of the object and of the object angle of the object.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG. 1 schematically shows a block diagram of a device according to one specific embodiment of the present invention.

[0022] FIG. 2 schematically shows a top view of a vehicle including a device and an object for explaining a possible collision.

[0023] FIG. 3 shows dependencies of the relative velocity on the vehicle velocity for various object velocities of the object.

[0024] FIG. 4 shows relative velocities and object angles of an object as a function of the object velocity and of the intersecting angle between the respective trajectories.

[0025] FIG. 5 shows a flow chart of a method for detecting critical transverse movements according to one specific embodiment of the present invention.

[0026] In all figures, identical or functionally identical elements and devices are provided with the same reference numerals.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0027] FIG. 1 shows a block diagram of a device 1 for detecting critical transverse movements, in accordance with an example embodiment of the present invention. Device 1 includes a first antenna device 21 having a narrow far range field of view 10 and a second antenna device 22 having a wide close-up range field of view 11. Antenna devices 21, 22 each include at least one transmitting antenna for emitting a radar signal and a plurality of receiver antennas for receiving the reflected radar signal. According to some specific embodiments, the antennas may be designed both for emitting radar signals as well as for receiving radar signals.

[0028] First antenna device 21 or second antenna device 22 may alternatively be operated via a switching device 23. A first control unit 24 may activate antenna device 21, 22 for emitting a FMCW-modulated radar signal and may evaluate the received FMCW modulated reflected radar signal.

[0029] A second control unit 25 is coupled to second antenna device 22 and activates this antenna device to emit a CW-modulated radar signal. Second control unit 25 is further designed to generate radar data based on the received CW-modulated radar signal and to transmit them to a computing device 3 for evaluation. Antenna devices 21, 22, switching device 23, first control unit 24 and second control unit 25 constitute a radar device 2 of device 1.

[0030] Computing device 3 includes a detection device 31, which initially ascertains collision-relevant spectral ranges of the radar data. For this purpose, detection device 31 receives the value of the ego velocity of radar device 1, determined by a sensor device not shown, which may correspond, in particular, to an ego velocity of a vehicle that includes radar device 1. The determination of the relevant spectral ranges is explained in greater detail with reference to the following figures.

[0031] FIG. 2 schematically shows a top view of a vehicle 8 including device 1, an object 4 being situated in a surroundings of vehicle 8. Object 4 may, for example, be a pedestrian, a bicyclist or another vehicle. Vehicle 8 moves at ego velocity v_ego along a straight vehicle trajectory 7, which defines an x-axis and a y-axis perpendicular thereto. The main radiation direction of radar device 2 also extends along the x-axis, however, the present invention is not limited to radar devices 2 aligned in this manner. Object 4 appears at an object angle or azimuth angle φ relative to the x-axis and moves at an object velocity v_obj along an object trajectory 6 toward a collision point 15. Object trajectory 6 and vehicle trajectory 7 intersect at collision point 15 at an intersecting angle ψ. To derive the measurable relative velocity v_rad of object 4, it is necessary to use the projection v_proj of object velocity v_obj on a connecting line 9 between device 1 and object 4.

[0032] In the event ego velocity v_ego of vehicle 8 and object velocity v_obj are selected in such a way that vehicle 8 is situated on a collision course with object 4, then neither relative velocity v_rad nor object angle φ change over time.

[0033] FIG. 3 illustrates the relative velocities of a perpendicularly intersecting object as a function of ego velocity v_ego of vehicle 8 for three different object velocities v_obj of 10 km/h, 15 km/h and 20 km/h.

[0034] Based on geometrical considerations, it is possible to derive the following correlation between relative velocity v_rad, of object velocity v_obj, of ego velocity v_ego of vehicle 8, object angle φ and intersecting angle ψ:

v.sub.rad=v.sub.obj cos(φ−ψ)+v.sub.ego(1−cos φ)

[0035] In the event object 4 and device 1 are situated on a collision course, it is also possible to derive the following relationships:

[00001]$\frac{-\sin \phi }{v_{\mathrm{obj}}}=\frac{\sin \left(90°+\phi +90°-\phi \right)}{v_{\mathrm{ego}}}$$\tan \phi =\frac{-\sin \psi }{\frac{v_{\mathrm{ego}}}{v_{\mathrm{obj}}}-\cos \psi }$

[0036] The above three relationships allow each intersecting angle ψ and each object velocity v_obj to be assigned exactly one collision-relevant relative velocity v_rad and one collision-relevant object angle φ using the firmly ascertained value of ego velocity v_ego.

[0037] FIG. 4, for example, illustrates the dependency of relative velocity v_rad and of object angle φ on intersecting angle  4 and of object velocity v_obj for an ego velocity v_ego of vehicle 8 of 25 km/h. For example, a perpendicularly intersecting trajectory, i.e., an intersecting angle ψ of 90 degrees at an object velocity v_obj of 15 km/h results in a relative velocity v_rad of −1.2 meters and an object angle φ of −31 degrees.

[0038] The collision-critical range is preferably expanded with the aid of tolerance ranges, so that the finite expansions of vehicle 8 and of object 4 are taken into consideration. In this case, it may be taken into consideration according to one specific embodiment that deviations in longer distances have lesser of an influence than in the case of shorter distances. Thus, the tolerance ranges and the collision-critical ranges may be a function of distance.

[0039] Returning to FIG. 1, detection unit 31 may limit the spectral range to be evaluated to those ranges that correspond to objects 4 having a negative relative velocity v_rad, since only such objects 4 are able to collide in the near future with vehicle 8.

[0040] Detection unit 31 may also limit the spectral range to relative velocities v_rad and object angles φ that satisfy the dependency discussed above. Object velocities v_obj and intersecting angles ψ remaining as parameters may be selected in particular, in predefined ranges. For example, it is possible to take only intersecting angles ψ into consideration that are smaller than predefined threshold values. Furthermore, objects 4 having object velocities v_obj, which are lower than a predefined threshold value, may be disregarded. Only objects 4, whose object velocity v_obj exceeds the predefined threshold value, are detected and evaluated.

[0041] Detection unit 31 detects objects within the limited collision-relevant spectral ranges of the radar data. While the evaluation below is described for one single object 4, detection unit 31 may also detect a plurality of objects 4 in parallel. Detection unit 31 ascertains relative velocity v_rad and object angle φ of object 4, in particular, based on the radar data. Multiple signal components may, for example, be detected as peaks in the spectrum and object angles φ may be determined based on separate DML angle estimations (deterministic maximum likelihood).

[0042] In the event detection unit 31 has detected an object 4, a tracking unit 32 ascertains a temporal course of relative velocity v_rad and of object angle φ of object 4.

[0043] An evaluation unit 33 is designed to determine whether the values of relative velocity v_rad and of object angle φ remain essentially constant. For this purpose, evaluation unit 33 may investigate whether an amount of a gradient of relative velocity v_rad and of the incident angle does not exceed respective threshold values. More generally, evaluation unit 33 is able to ascertain a degree of criticality or likelihood that relative velocity v_rad and object angle φ remain essentially constant, i.e., a collision with object 4 takes place. The degree of criticality may, in particular, be iteratively adapted based on new radar data.

[0044] Evaluation unit 33 may activate switching device 23 as a function of the calculated degree of criticality in order to select an appropriate antenna of antenna devices 21, 22.

[0045] Device 1 further includes a first calculating device 12 and a second calculating device 14. First calculating device 12 is designed to more precisely calculate and track a velocity, a distance and an object angle φ of object 4 based on the FMCW radar data. Second calculating device 14 is designed to carry out a pedestrian recognition based on the CW radar data by calculating a micro-Doppler effect. The results of the pedestrian recognition may be taken into consideration by first calculating device 12.

[0046] The device further includes an emergency braking unit 13, which is able to carry out an emergency braking as a function of the data ascertained by first calculating device 12. For this purpose, the degree of criticality generated by calculating device 33 may also be taken into consideration as a check for plausibility.

[0047] FIG. 5 illustrates a flow chart of a method for detecting critical transverse movements. The method may be carried out, in particular, by a device 1 described above.

[0048] In a method step S1, a CW radar signal is emitted and the reflected CW radar signal is received and radar data are generated. Radar device 2 used for this purpose may, as described above, include multiple antenna devices 21, 22 having different detection ranges 10, 11.

[0049] In method step S2, collision-relevant spectral ranges of the radar data are ascertained as a function of ego velocity v_ego of radar device 2. Relative velocity v_rad may, in particular, be limited to negative ranges. As described above, values or value ranges of relative velocities v_rad and corresponding object angles φ to be detected may also be limited, so that merely collision-critical situations are taken into consideration.

[0050] In method step S3, an object 4 is detected based on the collision-relevant spectral ranges of the radar data, and a time dependency of relative velocity v_rad and of object angle φ of object 4 is also ascertained.

[0051] In method step S4, a critical transverse movement of object 4 is detected using the time dependency of relative velocity v_rad and of object angle φ of object 4.

[0052] In an optional further method step S5, particular antenna devices 21, 22 may be selected or activated as a function of the result of the detection. In addition, an implementation of an emergency braking is possible, as described above.