METHOD AND DEVICE FOR OPERATING ULTRASONIC SENSORS OF A VEHICLE

Abstract

A method for operating ultrasonic sensors of a vehicle. In the method, front ultrasonic sensors of the vehicle are used to detect wind noise at the vehicle, rear ultrasonic sensors of the vehicle are used to detect a road condition in the area of the vehicle, and lateral ultrasonic sensors of the vehicle are used to detect objects in the area of the vehicle.

Claims

1-10. (canceled)

11. A method for operating ultrasonic sensors of a vehicle, comprising the following steps: detecting, using front ultrasonic sensors of the vehicle, wind noises at the vehicle; detecting, using rear ultrasonic sensors of the vehicle, a road condition in an area of the vehicle; and detecting, using lateral ultrasonic sensors of the vehicle, objects in the area of the vehicle.

12. The method as recited in claim 11, wherein noise levels detected at the ultrasonic sensors are evaluated to detect the wind noises and/or to detect the road condition and/or to detect the objects.

13. The method as recited in claim 11, wherein echoes detected at the ultrasonic sensors are evaluated to detect the objects.

14. The method as recited in claim 11, wherein wind noises and/or objects mapped in pieces of sensor information of those of the ultrasonic sensors oriented to a rear of the vehicle are compensated for using the detected wind noises and/or the detected objects.

15. The method as recited in claim 11, wherein wind noises and/or road conditions mapped in pieces of sensor information of those of the ultrasonic sensors oriented to the side are compensated for using the detected wind noises and/or the detected road condition.

16. The method as recited in claim 11, wherein pieces of sensor information of the ultrasonic sensors situated in pairs symmetrically with respect to a vehicle longitudinal axis of the vehicle are evaluated together.

17. The method as recited in claim 16, wherein the pieces of sensor information of the ultrasonic sensors situated on one vehicle side are used to detect passing vehicles or passed other vehicles on the vehicle side.

18. A device configured to operate ultrasonic sensors of a vehicle, the device configured to: detect, using front ultrasonic sensors of the vehicle, wind noises at the vehicle; detect, using rear ultrasonic sensors of the vehicle, a road condition in an area of the vehicle; and detect, using lateral ultrasonic sensors of the vehicle, objects in the area of the vehicle.

19. A non-transitory machine-readable memory medium on which is stored a computer program for operating ultrasonic sensors of a vehicle, the computer program, when executed by a computer, causing the computer to perform the following steps: detecting, using front ultrasonic sensors of the vehicle, wind noises at the vehicle; detecting, using rear ultrasonic sensors of the vehicle, a road condition in an area of the vehicle; and detecting, using lateral ultrasonic sensors of the vehicle, objects in the area of the vehicle.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Specific embodiments of the present invention are described hereinafter with reference to the FIGURES, with neither the FIGURES nor the description to be interpreted as restricting the present invention.

[0026] FIG. 1 shows a depiction of a vehicle having differently oriented ultrasonic sensors and a device according to one exemplary embodiment of the present invention.

[0027] The FIGURE is merely schematic and is not true to scale. Identical reference numerals identify identical or identically acting features in the FIGURE.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

[0028] FIG. 1 shows a depiction of a vehicle 100 having differently oriented ultrasonic sensors 102 and a device 104 according to one exemplary embodiment. Ultrasonic sensors 102 are situated distributed over vehicle 100. Ultrasonic sensors 102 are consecutively numbered from one through thirteen clockwise here.

[0029] First ultrasonic sensor 102 is situated here at a left front corner of vehicle 100 and is oriented to the left with respect to a vehicle longitudinal axis 106 of vehicle 100. Second ultrasonic sensor 102 is also situated on the left front corner and is oriented forward and diagonally to the left with respect to vehicle longitudinal axis 106. Third ultrasonic sensor 102 is also situated on the left front corner and oriented forward.

[0030] Fourth ultrasonic sensor 102 is situated on a right front corner of vehicle 100, symmetrically in relation to third ultrasonic sensor 102 with respect to vehicle longitudinal axis 106, and is oriented forward like third ultrasonic sensor 102. Fifth ultrasonic sensor 102 is also situated on the right front corner, symmetrically in relation to second ultrasonic sensor 102 with respect to vehicle longitudinal axis 106 and is oriented forward and diagonally to the right with respect to vehicle longitudinal axis 106. Sixth ultrasonic sensor 102 is also situated on the right front corner, symmetrically in relation to first ultrasonic sensor 102 with respect to vehicle longitudinal axis 106, and is oriented to the right with respect to vehicle longitudinal axis 106.

[0031] The number seven is not assigned.

[0032] Eighth ultrasonic sensor 102 is situated on a right rear corner of vehicle 100 and is oriented to the right with respect to vehicle longitudinal axis 106. Ninth ultrasonic sensor 102 is also situated on the right rear corner and is oriented to the rear and diagonally to the right with respect to vehicle longitudinal axis 106. Tenth ultrasonic sensor 102 is also situated on the right rear corner and oriented to the rear.

[0033] Eleventh ultrasonic sensor 102 is situated on a left rear corner of vehicle 100, symmetrically to tenth ultrasonic sensor 102 with respect to vehicle longitudinal axis 106, and is oriented to the rear like tenth ultrasonic sensor 102. Twelfth ultrasonic sensor 102 is also situated on the left rear corner, symmetrically to ninth ultrasonic sensor 102 with respect to vehicle longitudinal axis 106, and oriented to the rear and diagonally to the left with respect to vehicle longitudinal axis 106. Thirteenth ultrasonic sensor 102 is also situated on the left rear corner, symmetrically to eighth ultrasonic sensor 102 with respect to vehicle longitudinal axis 106, and is oriented to the left with respect to vehicle longitudinal axis 106.

[0034] Each ultrasonic sensor 102 may carry out echolocation of objects in its particular detection area 108 using emitted ultrasound and map a distance to the object in a piece of sensor information 110. Alternatively or additionally, each ultrasonic sensor 102 may detect ambient noises and map them in sensor information 110. An intensity of the ambient noises is mapped in each case in a noise level 112 of sensor information 110.

[0035] Pieces of sensor information 110 of all ultrasonic sensors 102 are read in by device 104.

[0036] When a vehicle 100 drives slowly, for example, during maneuvering, parking, or in a traffic jam, the echolocation functions as intended and few ambient noises are detected. If the ambient noises become louder than the echoes of the ultrasound, the echolocation only still functions to a limited extent.

[0037] For example, airflow at a body of vehicle 100 generates a wind noise, which is detected by ultrasonic sensors 102. Furthermore, tires of vehicle 100 generate a rolling noise, which is also detected by ultrasonic sensors 102. If a roadway is damp or wet, the tires additionally generate a water noise or wet hissing, which is also detected by ultrasonic sensors 102. The wind noise, the rolling noise, and the wet hissing are included with external noises of other noise sources in the ambient noise. At least the wind noise, the rolling noise, and the water noise become louder with increasing speed of vehicle 100.

[0038] In accordance with the exm, pieces of sensor information 110 from ultrasonic sensors 102 oriented essentially forward are used to detect the wind noise. Pieces of sensor information 110 from ultrasonic sensors 102 oriented essentially to the rear are used to detect the water noise and the rolling noise. Pieces of sensor information 110 of ultrasonic sensors 102 oriented essentially to the side are used to detect external noises of other noise sources.

[0039] In other words, a method is presented for sensor selection in the field of wetness detection on the roadway with the aid of ultrasound.

[0040] Presently, the roadway wetness or the water column specification in millimeters on a roadway may not be directly measured. A wet roadway may be inferred indirectly from various operating states of the vehicle. This may take place, for example, due to the windshield wiper activity or ESP interventions. A continuous measurement of the roadway condition regarding moisture presently does not exist.

[0041] For obstacle detection, ultrasonic sensors are attached in the vicinity of the wheel cases. One significant problem in the use of obstacle detection during faster travel are the driving noises, which are then overlaid on the echo emitted by the sensors and thus sometimes severely restrict the distance measurement. The more water from the tires sprays against the wheel cases, the louder the driving noise is and the more severe the restriction is. The noise level primarily reaches the sensor directly via the air, but may also be received by the sensor indirectly via structure-borne noise. These noises may be computed as the noise level or as noise (disturbance variable, noise value) in the ultrasound control unit. The noise level may be output via CAN to other control units in the vehicle.

[0042] Driving tests have shown that the quality of the detection of roadway properties, for example, a wetness detection or a detection of ambient properties, for example, the detection of other vehicles, is strongly dependent on the particular selection of the involved ultrasonic sensors (USS). Not every sensor position (presently up to 12 sensors may be installed per vehicle) is equally well-suited or it may even have a negative influence with respect to the particular detection method. Therefore, a suitable sensor selection plays a very large role in the ability to make robust and high-quality statements with respect to wetness on the roadway.

[0043] Various ambient or roadway properties may be detected with the aid of the ultrasonic sensors. These are object detection, wetness detection on the roadway, and wind detection.

[0044] In accordance with an example embodiment of the present invention, the wind speed in the longitudinal direction is computed using the noise levels of the front sensors, objects are detected using the noise levels of the lateral sensors, and wetness on the roadway is measured using the noise levels of the rear sensors. Since wind and objects also influence the noise level of the rear sensors, the measurement of the wetness is corrected with the aid of the pieces of wind and object information.

[0045] The four laterally situated sensors (numbers 1, 6, 8, 13) are decisive for a detection of objects (other road users, for example, vehicles, trucks, . . . ). The particular difference signal of the noise values of the front sensors (numbers 1, 6) and the rear sensors (numbers 13, 8) is evaluated.

U diff. 1,6 [mV]=U 1 [mV]U 6 [mV]

[0046] U diff. 1, 6>threshold object is a vehicle falling behind another vehicle on the left vehicle side U diff. 1, 6<-threshold object is a vehicle passing the ego vehicle on the right vehicle side.

[0047] U diff. 13, 8 behaves in the same way.

[0048] In addition, a falling behind or passing process may be concluded via the behavior with respect to time of difference signals U diff. 1, 6 and U diff. 13, 8. During a passing process, an increase of the difference signal first occurs in a vehicle-specific manner on the right side in the travel direction at the front lateral sensors (1, 6), chronologically thereafter corresponding to the difference speed at the rear sensors (8, 13). In the case of a falling-behind process, it is correspondingly the other way around and on the left side. Furthermore, a continuous object (guide rail, wall, . . . ) and its location (distance, left or right) may be concluded with a chronologically (longer) constant difference signal.

[0049] For this purpose, various threshold values are used for various types of objects (automobile, truck, . . . ). An inference of the particular object may thus also be drawn during driving operation and validated, for example, using the objects from the radar/LIDAR/camera surroundings. For object detection, the sensors (1, 6, 8, 13) may also emit active ultrasonic signals in the entire feature-specific speed band (presently >=60 km/h) in addition to measuring the noise level and may detect objects with the aid of the received echoes, as long as they are not suppressed by excessively high noise levels.

[0050] The four rear sensors (numbers 9, 10, 11, 12) are decisive for a detection of the roadway wetness. The best measurement results with respect to the presently existing roadway wetness may be achieved using this sensor selection. At these sensor positions, for example, the wind influence is least.

[0051] If no object is detected in the detection area by the vehicle-specific object detection, all rear sensors (numbers 9, 10, 11, 12) may be used for wetness detection. The sensors may be operated actively or also inactively for this purpose, since the noise value may be ascertained and processed in any case.

[0052] If a continuous object (guide rail, etc.) is detected in the detection area by the vehicle-specific object detection (for example, also using radar, camera, or LIDAR), the wetness detection may be suspended. Alternatively, all rear sensors (numbers 9, 10, 11, 12) may nonetheless be incorporated for the wetness detection. However, the sensor values of the rear sensors are reduced by a sensor-specific and possibly object-specific correction value. The object influence on the sensor-specific noise level is thus compensated for.

[0053] If a short-term object (vehicle, etc.) is detected in the detection area by the vehicle-specific object detection, the rear sensors (numbers 9, 10, 11, 12) are partially deactivated for wetness detection in an object-specific manner. Object groups are formed in this case, which have a similar influencing pattern (profile and intensity) with respect to the noise level of a single sensor. In accordance with each object group, it is defined which sensors for wetness detection may still supply a contribution to the wetness detection during the influence.

[0054] The noise level of all other sensors is not taken into consideration for the computation of the wetness during the time of this influence. For example, in the case of an object type automobile, only the two sensors close to the object are deactivated (see also continuous object). In the case of an object type truck, in contrast, all rear sensors are not taken into consideration, since presently wetness detection may not be carried out via ultrasonic sensors here.

[0055] The continuous noise value of the sensor raw signal is converted into a sensor-individual status. An index (sensor designation) is associated with each sensor:

[0056] i=[1 2 3 4 5 6 NaN 8 9 10 11 12 13], since the index 7 is not used.

[0057] The result is output as a status vector:

[0058] Z=[1 2 3 4 5], the status being interpreted as follows and intermediate values being able to occur:

[0059] 1: dry

[0060] 2: damp

[0061] 3: wet

[0062] 4: very wet

[0063] 5: risk of hydroplaning

[0064] n: support point of status vector Z

[0065] v: speed (dependence)

[0066] t: time (dependence)

[0067] s(i): sensor selection

[0068] s(i).SZ(t): sensor result at point in time t

[0069] stst noise: support point value of the noise value

[0070] tracker noise: present noise value of the sensor

[0071] and may be computed as follows:

[00001]$s\left(i\right).Math.\mathrm{SZ}\left(t\right)=\frac{{}_{n=1}^5\left(Z\left(n\right)*\frac{1}{{\left(s.Math.t.Math.s.Math.t_{n.Math.o.Math.i.Math.s.Math.e\left(i,n,v\right)}-t.Math.r.Math.a.Math.c.Math.k.Math.e.Math.r_{n.Math.o.Math.i.Math.s.Math.e}\left(i,t\right)\right)}^2}\right)}{{}_{n=1}^5\left(\frac{1}{{\left(s.Math.t.Math.s.Math.t_{n.Math.o.Math.\mathrm{ise}\left(i,n,v\right)}-t.Math.r.Math.a.Math.c.Math.k.Math.e.Math.r_{n.Math.o.Math.i.Math.s.Math.e}\left(i,t\right)\right)}^2}\right)}$

[0072] A sensor fusion value is computed across all sensors via a fusion factor from the individually computed sensor-individual results. The dimension of the sensor-individual fusion factor is dependent on the object detection.

[0073] Examples of various fusion factors k in the case of roadway detected as wet or very wet. The crossed-out sensors are not taken into consideration in each case.

[0074] k 1 (i)=[0 0 0 0 0 0 NaN 0 2 4 4 2 0] no object detected

[0075] k 2 (i)=[0 0 0 0 0 0 NaN 0 4 2 0] passing vehicle (car) detected

[0076] k 3 (i)=[0 0 0 0 0 0 NaN 0 0] passing vehicle (truck) detected

[0077] k 4 (i)=[0 0 0 0 0 0 NaN 0 2 4 0] vehicle falling behind (car) detected

[0078] Computation of the fused result:

[00002]${\mathrm{SZ}}_{f.Math.u.Math.s\left(t\right)}=\frac{{}_{i=1}^{1.Math.3}\left(s\left(i\right).Math.\mathrm{SZ}\left(t\right)*\frac{k\left(i\right)}{v_{{\mathrm{tracker}}_{n.Math.o.Math.i.Math.s.Math.e\left(t\right)}}}\right)}{{}_{i=1}^{1.Math.3}\left(\frac{k\left(i\right)}{v_{{\mathrm{tracker}}_{n.Math.o.Math.i.Math.s.Math.e\left(t\right)}}}\right)}$

[0079] SZ fus (t): fused result of the wetness detection

[0080] v_tracker noise: present vehicle speed

[0081] The four front sensors (2, 3, 4, 5) are decisive for a detection of the wind in the vehicle longitudinal direction. The effect is utilized for this purpose that the wind influence on the noise level of the individual sensors occurs more intensely in the front. This effect may be ascertained in an experimental trial and may be mapped in a corresponding model value. Airflow is applied to the vehicle in the longitudinal direction for this purpose (for example, in a wind tunnel). Windspeed-dependent sensor-individual noise values may be mapped therefrom and the following relationship may be formed:

V wind [km/h]=V airflow [km/h]V vehicle [km/h](1)

V vehicle=0 km/h (in the wind tunnel)(2)

V airflow [km/h]N sensor [mV]*k [km/h/mV](3)

V wind [km/h]=N sensor [mV]*k[km/h/mV]V vehicle [km/h](4)

[0082] V: speed

[0083] N: noise value noise

[0084] k: speed-dependent correction factor

[0085] or in road driving trial:

V wind [km/h]=V airflow [km/h]V vehicle [km/h](1)

V vehicle=0 km/h (windless day)(2)

V airflow [km/h]=V vehicle [km/h](3:

V airflow [km/h]N sensor [mV]*k [km/h/mV](4)

V wind [km/h]=N sensor [mV]*k[km/h/mV]V vehicle [km/h](5)

[0086] Since headwind increases the noise level of the four rear sensors in the same way and tailwind reduces it, the windspeed computed using the four front sensors is used to compensate for the noise level of the four rear sensors.

[0087] The example embodiment of the present invention may be used in any passenger vehicle having an integrated, automatic parking assistant as a software feature. The method may be used in principle in all vehicles having ultrasonic sensors. Since only an already computed signal is provided on the CAN bus and a warning is output to the driver due to this signal, a minimal implementation using software changes on the ultrasound control unit and on the HMI is possible very cost-effectively.

[0088] Finally, it is to be noted that terms such as having, including, etc. do not exclude other elements or steps and terms such as a or one do not exclude a plurality.