Method for identifying obstacles in a pivoting range of a motor vehicle door

Abstract

A method for identifying obstacles in a pivoting range of a motor vehicle door (4) comprises: a) identifying (S1) a number of object features (701-703) which each indicate a position of a supposed obstacle (50) in the surroundings (10) of the motor vehicle (1), by means of distance measurements using a number of ultrasonic sensors (13, 14), wherein the position (501) of an obstacle (50) measured by more than one of the ultrasonic sensors (13, 14) is identified by means of trilateration, and the position (502, 503) of an obstacle (50) measured by only one of the ultrasonic sensors (14) is determined in accordance with the field of view of the ultrasonic sensor (14) in question, and b) blanking out (S2) a non-trilaterated object feature (703) which has moved from a first region within a first distance (y1) from the door (4) into a second region between a second (y2) and third distance (y3) from the door (4).

Description

[0001] METHOD FOR DETERMINING OBSTACLES IN A SWINGING AREA OF A MOTOR VEHICLE DOOR

[0002] The present invention relates to the field of motor vehicle ultrasonic sensors and, in particular, to a method for determining obstacles in a pivoting range of a door of a motor vehicle by means of a number of ultrasonic sensors mounted on the motor vehicle.

[0003] To measure the surroundings of a motor vehicle using ultrasound, several ultrasonic sensors are mounted on the vehicle. These sensors transmit and receive ultrasonic signals into the vehicle's surroundings. Based on the signal propagation times between the transmission and reception of an ultrasonic signal, distances to reflection points on suspected obstacles in the vehicle's surroundings are determined. The determined distances can be represented as circular (if the transmitting and receiving ultrasonic sensors are identical) or elliptical (if the transmitting and receiving ultrasonic sensors are different ultrasonic sensors) loci of possible reflection points in the surroundings. By determining the intersection points of the loci, positions can be determined where an obstacle is suspected to be in the vehicle's surroundings. This process is called "trilateration."

[0004] A driver or passenger of a motor vehicle can unlock the vehicle by operating a remote control integrated into the vehicle key. With keyless entry, a motor vehicle can also unlock automatically when a driver or passenger carrying a key equipped with an RFID transponder approaches the vehicle. To increase convenience, once the vehicle has been unlocked, a door of the vehicle can be automatically opened for the driver or passenger. To signal to the vehicle which of several doors should be opened, the passenger can, for example, touch a door opener button attached to the door handle of the respective door, whereupon the respective door will open automatically. In addition to a respective door opener button, the ultrasonic sensors can also be integrated into the door handle of the respective door.

[0005] In the described application, however, a person approaching the door to press the door release button comes so close to the motor vehicle that they are only seen by one ultrasonic sensor or only by several ultrasonic sensors installed at essentially the same position along the longitudinal direction of the motor vehicle. For example, the person standing close to the motor vehicle is only seen by the ultrasonic sensor in the relevant door handle. At the same time, however, the person is outside the field of view of ultrasonic sensors at other positions along the longitudinal direction of the motor vehicle, such as the ultrasonic sensor in the other door handle. Thus, although the distance to the person can still be determined, their position can no longer be trilaterated.

[0006] In practice, this leads to the problem that the door of the motor vehicle is not opened or is not opened reliably after the person has touched the door opener button and stepped back in anticipation of the door opening automatically.

[0007] Against this background, the object of the invention is to improve a method for determining obstacles in a pivoting area of ​​a door of a motor vehicle by means of ultrasound.

[0008] According to a first aspect, a method for determining obstacles in a pivoting range of a door of a motor vehicle using a number of ultrasonic sensors mounted on the motor vehicle is proposed. The method comprises: a) determining a number of object features, each indicating a position of a suspected obstacle in the surroundings of the motor vehicle, by means of distance measurements using a number of ultrasonic sensors of the motor vehicle, wherein the position of an obstacle measured by more than one of the ultrasonic sensors is determined by trilateration using the respective measured distances and the positions of the respective ultrasonic sensors, and the object feature in question is marked as trilaterated, and the position of an obstacle measured by only one of the ultrasonic sensors,is determined according to the measured distance and a field of view of the respective ultrasonic sensor and the respective object feature is marked as non-trilaterated, and b) masking out a non-trilaterated object feature that has moved from a first area within a first distance to the door to a second area between a second and third distance to the door, wherein the first distance is smaller than the second distance and the second distance is smaller than the third distance.

[0009] The proposed solution is based on the finding that in the described application, the door is not opened because the position of the person who touched the door release button and then stepped back to the side cannot be trilaterated and it is wrongly assumed that the person is not located to the side of the door's swing range at the measured distance, but is still within the door's swing range.The proposed solution is further based on the idea that under certain precisely defined conditions relating to a person's movement pattern, in particular movement from within the first distance to the area between a second and third distance from the door, it can be recognized that the obstacle is the person requesting the door to be opened. Accordingly, an object feature representing this person need not be considered an obstacle and can be masked out, thus allowing the door to be opened. In contrast, object features that do not meet the defined conditions are not masked out, so that other obstacles continue to be correctly detected.Accordingly, the proposed solution advantageously enables the door to be opened in response to the actuation of the door opener button by masking out the object feature representing the person, while at the same time collisions of the door with other obstacles can be avoided.

[0010] The distance measurements in step a) may include: transmitting a number of ultrasonic signals into the surroundings of the motor vehicle; receiving a number of reflected ultrasonic signals from the surroundings of the motor vehicle; determining a respective distance to the suspected obstacle based on a signal propagation time between the transmission and reception of the respective ultrasonic signal. The trilateration in step a) may include: determining several circular or elliptical loci of possible reflection points in the surroundings of the motor vehicle using the measured distances and the positions of the respective ultrasonic sensors involved; and determining an intersection point of the determined loci as the position of the obstacle.Determining the position according to a field of view of the ultrasonic sensor in step a) in the event that trilateration is not possible because the obstacle was only measured by one of the ultrasonic sensors may include: determining a single circular locus of possible reflection points in the surroundings of the motor vehicle using the measured distance as the radius and the position of the ultrasonic sensor as the center point; and determining an intersection point of the determined locus with a sensor axis of the ultrasonic sensor. A sensor axis is, for example, the line in the field of view of the ultrasonic sensor on which the radiation intensity is maximum.

[0011] A respective object feature can be understood as a data structure comprising several entries, whereby at least one entry of the data structure contains the specific position of the suspected obstacle.

[0012] The determined position is, in particular, a two-dimensional position of the suspected obstacle in a projection from above onto the road plane. Unless otherwise stated or clear from the context, the term "number" always means a number of one or more elements, i.e., a number N > 1.

[0013] According to one embodiment, steps a) and b) are carried out cyclically, and in step b) the object feature is only hidden if at least a predetermined time ti has elapsed before the object feature appears in the first area and no object features have been determined at least in the first area and in the second area over a predetermined number of runs of step a).

[0014] Preferably, the condition is that over a predetermined number of runs of step a), no object features were determined in the entire search window described later.

[0015] Accordingly, it is first ensured that a static situation exists in which there are no static obstacles and no other dynamic obstacles such as moving branches, stray animals and the like in the surroundings of the motor vehicle or at least in the considered areas of the surroundings of the motor vehicle, before a detection is carried out as to whether a person has moved towards the door and then moved from the first to the second area.

[0016] In this way, object features that actually indicate the movement of the person with the desire to open the door can advantageously be distinguished even better from other object features, and the masking of the object feature that indicates the person who has stepped aside can be done even more precisely.

[0017] According to a further embodiment, step b) is only performed for non-trilaterated object features in the second region if a non-trilaterated object feature is previously detected in the first region that has a velocity whose magnitude exceeds a predetermined threshold. Object features with a velocity above the predetermined threshold are also referred to as dynamic object features.

[0018] By only masking out object features in the second area that can be assigned to a dynamic object feature previously detected in the first area through a temporal relationship, the probability is increased that the object feature to be masked out is actually an object feature that identifies the person who touched the door opener button and then stepped aside.

[0019] Thus, the safety against collisions with other obstacles can be advantageously increased with the proposed method.

[0020] It should be noted that a speed of an object feature can be determined by determining the object features according to step a) at a first time, determining the object features again according to step a) at a second time, and comparing the determined object features of the two times to determine which of the object features have remained in place (speed zero) and which have moved over what distance (speed equal to distance traveled divided by the time interval between the times).

[0021] According to a further embodiment, in step b) the object feature is only hidden if less than a predetermined time t2 has elapsed since the object feature left the first area within the first distance.

[0022] Accordingly, a temporal connection is advantageously established between the leaving of the first area in front of the door opener button and the entry of the object feature into the second area, ie the presumed stepping back of the person to the side, and the object feature is only hidden if such a temporal connection exists.

[0023] This advantageously makes it possible to define even more precisely when it is safe to mask out the object feature and when it may not be masked. According to a further embodiment, in step b), only object features located within a predefined search window are processed, which extends in the vehicle's longitudinal direction from a position in front of a position of a door handle of a door of the motor vehicle to a position at the rear end of the door's pivot range in the vehicle's longitudinal direction, and extends in the vehicle's transverse direction from in front of the first distance from the motor vehicle to the third distance from the motor vehicle.

[0024] By using a search window defined in this way, it is advantageously easier to track the movement of the object feature into the first area and from there into the second area, since generally no other object features are measured within the search window and thus an assignment of the object feature in question over several measuring cycles is particularly easy.

[0025] On the other hand, it also advantageously prevents object features at other positions outside the search window from being mistakenly identified as object features representing the person attempting to open the door. Thus, the precision of the proposed method can be further increased.

[0026] According to a further embodiment, the method further comprises: c) automatically opening the door of the motor vehicle in response to actuation of a door opener button attached to the door depending on the determined and non-masked object features located within the pivoting range of the door.

[0027] The method can therefore also be used as a method for automatically opening a door of a

[0028] Motor vehicle in response to the actuation of a door opener button. According to the described features, according to which the object feature representing the person is masked out upon recognition of a typical movement pattern of a person wishing to open the door, the door can advantageously be opened even if the user has actuated the door opener button and stepped aside, but due to failed trilateration of their position, is still incorrectly detected as standing in front of the door and blocking it.

[0029] According to a further embodiment, step c) comprises opening the door only if there are no object features in the pivoting range of the door, or in step c) a maximum opening angle for the door is determined based on the object features located within the pivoting range of the door and the door is only opened up to the determined maximum opening angle and this only if the determined maximum opening angle is greater than a predetermined minimum opening angle.

[0030] The specified minimum opening angle can be set appropriately for the vehicle model, allowing entry for a human driver or passenger starting at this angle. In particular, tests can be conducted and a minimum opening angle selected at which test subjects perceive entry as comfortable. For an exemplary vehicle model, the minimum opening angle is 26°, for example, but there is no restriction on a specific value for the minimum opening angle.

[0031] According to a second aspect, a computer program product is proposed which comprises instructions which, when the program is executed by a computer, cause the computer to carry out the method according to the first aspect or one of its embodiments.

[0032] A computer program product, such as a computer program means, can be provided or delivered, for example, as a storage medium, such as a memory card, USB stick, CD-ROM, DVD, or in the form of a downloadable file from a server in a network. This can be done, for example, in a wireless communications network by transmitting a corresponding file with the computer program product or the computer program means.

[0033] The computer can in particular be, for example, a vehicle control unit (ECU, Electronic Control Unit).

[0034] According to a third aspect, a control device for a motor vehicle is proposed for determining obstacles within a pivoting range of a door of the motor vehicle using a number of ultrasonic sensors mounted on the motor vehicle. The control device comprises: a first unit configured to determine a number of object features, each indicating a position of a suspected obstacle in the surroundings of the motor vehicle, by means of distance measurements using a number of ultrasonic sensors of the motor vehicle, wherein the position of an obstacle measured by more than one of the ultrasonic sensors is determined by trilateration using the respective measured distances and the positions of the respective ultrasonic sensors, and the respective object feature is marked as trilaterated, and the position of an obstacle measured by only one of the ultrasonic sensorsis determined according to the measured distance and a field of view of the respective ultrasonic sensor and the respective object feature is marked as non-trilaterated, and b) a second unit configured to mask out a non-trilaterated object feature that has moved from a first area within a first distance to the door to a second area between a second and third distance to the door, wherein the first distance is smaller than the second distance and the second distance is smaller than the third distance.

[0035] The respective unit can be implemented in hardware and / or software. In a hardware implementation, the respective unit can be embodied, for example, as a computer or a microprocessor. In a software implementation, the respective unit can be embodied as a computer program product, as a function, as a routine, as an algorithm, as part of a program code, or as an executable object. Furthermore, each of the units mentioned here can also be embodied as part of a higher-level control system of the motor vehicle, such as a central electronic control device and / or an engine control unit (ECU).

[0036] Furthermore, a motor vehicle with a control device according to the third aspect is proposed.

[0037] The embodiments, advantages and features described for the proposed method of the first aspect apply accordingly to the proposed computer program product of the second aspect, the proposed control device of the third aspect and the proposed motor vehicle.

[0038] Further possible implementations of the invention also include combinations of features or embodiments described above or below with respect to the exemplary embodiments that are not explicitly mentioned. In this case, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

[0039] Further advantageous embodiments and aspects of the invention are the subject of the dependent claims and the exemplary embodiments of the invention described below. The invention is explained in more detail below using preferred exemplary embodiments with reference to the accompanying figures.

[0040] Fig. 1 shows a motor vehicle with several ultrasonic sensors and a control device according to embodiments;

[0041] Fig. 2 illustrates the trilateration of a point-like object feature; Fig. 3 illustrates a situation in which trilateration is not possible;

[0042] Fig. 4 shows steps of a method for determining obstacles in the pivoting range of a door of the motor vehicle from Fig. 1 according to a first embodiment;

[0043] Fig. 5 illustrates the method according to the first embodiment using a plan view of a lateral environment of the motor vehicle;

[0044] Fig. 6 illustrates the method according to a second embodiment using a plan view of a lateral environment of the motor vehicle; and

[0045] Fig. 7 illustrates details of the step of masking an object feature according to a third embodiment.

[0046] In the figures, identical or functionally equivalent elements are provided with the same reference numerals unless otherwise indicated. In the figures, arrows labeled "F" indicate the vehicle's longitudinal direction and point toward the front of the vehicle.

[0047] Fig. 1 shows a motor vehicle 1 with several ultrasonic sensors 11-14 and a control device 2 according to embodiments.

[0048] In the illustrated embodiment, motor vehicle 1 has a front door 3 and a rear door 4. An ultrasonic sensor 13 is installed in a door handle 31 of the front door 3, and an ultrasonic sensor 14 is installed in a door handle 41 of the second door. Optionally, additional ultrasonic sensors can be installed, such as the ultrasonic sensor 12 installed in a side sill 5 of motor vehicle 1 or the ultrasonic sensor 11 installed in a front bumper 6 of motor vehicle 1. Furthermore, a respective door opener button 32, 42 is installed in the respective door handles 31, 41.

[0049] A driver or passenger of the motor vehicle 1, approaching the motor vehicle 1 with the intention of boarding, touches one of the door opener buttons 32, 42 to give a command to open one of the doors 3, 4. The command is received by a third unit 23 of the control device 2, which then controls the automatic opening of the respective door 3, 4.

[0050] It is desirable to avoid collisions between the automatically opening door 3, 4 and the driver or passenger or with other obstacles in the lateral environment of the motor vehicle 1. Therefore, the first unit 21 of the control device 2 uses the ultrasonic sensors 101-106 to measure the lateral environment of the 110 (Fig. 2) of the motor vehicle 1 and to determine object features of suspected obstacles in the environment of the motor vehicle 1. The second unit 22 performs post-processing of object features determined by the first unit 21, which will be discussed in more detail later. The third unit 23 controls the automatic opening of the respective door 3, 4 depending on the determined object features.

[0051] In the following, some measurement principles for measuring the surroundings of the motor vehicle 1 using ultrasound by the first unit 21 of the control device 2 are explained with reference to Fig. 2 and Fig. 3. These measurement principles are compatible with all the embodiments described below.

[0052] Fig. 2 shows a schematic top view of the motor vehicle 1 and a lateral environment 10 of the motor vehicle 1 and illustrates the trilateration of a point-shaped object feature 71. A point-shaped obstacle 51, such as a post, a tree trunk, or a person, is located in the environment 10 of the motor vehicle 1. It should be noted at this point that "point-shaped" here and throughout the present description encompasses not only the strictly mathematical meaning of a point, but also essentially encompasses point-shaped objects and object features that, in the mathematical sense, represent circles with a given radius that is small compared to the dimensions of the motor vehicle 1.

[0053] A first ultrasonic sensor 11 transmits an ultrasonic signal into the environment 10 of the motor vehicle 1, but receives no echo, at least in the configuration shown in Fig. 2, since the only obstacle 51 lies outside the field of view of the first ultrasonic sensor 11. A second ultrasonic sensor 12 transmits an ultrasonic signal into the environment 10 of the motor vehicle 1. The ultrasonic signal is reflected at a reflection point on the obstacle 51 and travels back to the ultrasonic sensor 12. Based on a propagation time difference between the transmission and reception of the ultrasonic signal, a distance from the reflection point at which the ultrasonic signal was reflected to the ultrasonic sensor 12 is determined. It is thus determined that possible reflection points lie on a circular locus 62 (only partially shown in Fig. 2) whose center is the position of the ultrasonic sensor 12.In the same way, the ultrasonic sensors 13 and 14 transmit and receive respective ultrasonic signals, and further circular loci 63, 64 of possible reflection points are determined. It is also conceivable that one of the ultrasonic sensors 11, 12, 13 transmits an ultrasonic signal and another of the ultrasonic sensors 11, 12, 13 receives the reflected ultrasonic signal. In this case, too, a locus of possible reflection points can be determined, which is represented as an ellipse, at whose focal points the participating ultrasonic sensors 12, 13 and / or 14 are located.

[0054] As can be seen in Fig. 2, the loci 62, 63, 64 approximately intersect at a single point. The first unit 21 (Fig. 1) therefore generates a point-like object feature 71, which indicates a position of the intersection point of the loci 62, 63, 64. The position can be approximately determined, for example, using a Gauss-Newton method. As can be seen in Fig. 2, the object feature 71 is located in front of the obstacle 51 in the direction of the motor vehicle 1; however, this is not critical for all practical applications and provides a small additional safety margin. The object feature 71 is also marked as a trilaterated object feature. This can be done, for example, by setting a flag indicating that the object feature 71 has been successfully trilaterated to "true."Alternatively, a blur measure, which may also be included in the object feature, can be set to a low value, for example to the mean square error resulting from the Gauss-Newton method, and the value of the blur measure can then be used to determine that the object feature is a trilaterated object feature.

[0055] Fig. 3 shows a schematic plan view of the motor vehicle 1 and a lateral environment 10 of the motor vehicle 1 and illustrates a situation in which trilateration is not possible.

[0056] In Fig. 3, a point-like obstacle 52 is located very close to the motor vehicle 1. The point-like obstacle 52 is only seen by the ultrasonic sensor 14; however, it is not located in the field of view of the ultrasonic sensors 11, 12, and 13, i.e., it is located outside the signal lobes emitted by the ultrasonic sensors 11, 12, and 13. Accordingly, during the measurement, only a single distance of a reflection point from the ultrasonic sensor 12 is determined, and only a single circular locus 54 is obtained. Accordingly, an exact position of the obstacle 52 cannot be determined. The position of the object feature 72 generated in this case is instead selected, for example, at the intersection point of the locus 54 with a sensor axis of the ultrasonic sensor 14. An object feature 72 obtained in this way is characterized as a non-trilaterated object feature.Again, the marking can be done by setting a flag, for example, to "false," or by setting a blur measure to a large value. For example, the blur measure can be set to the entire width of the signal beam emitted by the ultrasonic sensor 102 at the measured distance.

[0057] It has been described how a position of a particular object feature 71, 72 can be determined by trilaterizing multiple measurements with multiple ultrasonic sensors 11-14 or by selecting according to a field of view, such as a sensor axis (center axis of the field of view), of a single ultrasonic sensor 14.

[0058] Furthermore, it is also conceivable to determine a speed of the suspected obstacle (51, 52) and to store it in the object feature 71, 72. For this purpose, the measurement is carried out as described at a first time t1 and at a second time t2, and the positions of the determined object features 71, 72 at the two times t1, t2 are compared. Ontology-based forecasting methods can also be used here, wherein a forecast of the positions at the second time t2 is determined on the basis of the positions determined during the measurement at the first time t1, and the forecast is compared with the actually determined positions at time t2 in order to achieve an assignment between object features 71, 72 determined at time t1 and object features 71, 72 determined at time t2.A velocity vector is then added to the object features 71, 72 determined at time t2 as a velocity specification, which describes the movement from an object feature 71, 72 determined at time t1 to the corresponding object feature 71, 72 determined at time t2. If a velocity specification, i.e., an absolute value of the velocity vector, is greater than a predefined threshold value, the corresponding object feature 71, 72 can be marked as dynamic. If it is less than or equal to the predefined threshold value, the corresponding object feature 71, 72 can be marked as static. This marking can again be done via a flag or via the value of the velocity specification.

[0059] Fig. 4 shows steps of a method for determining obstacles in the pivoting range of a door of the motor vehicle 1 from Fig. 1 according to a first exemplary embodiment, and Fig. 5 illustrates the method according to the first exemplary embodiment using a plan view of a lateral environment 10 of the motor vehicle 1. The first exemplary embodiment is described with reference to Figs. 1, 4 and 5. Fig. 5 shows a left side of the motor vehicle 1 with the door handles 41, 31 and the door opener buttons 42, 32 and the ultrasonic sensors 14, 13 installed therein. The description below focuses on the rear door 4, which can be rotated about a pivot point 44 and has a pivoting range 40 that should be free of obstacles so that the door 4 can be opened automatically.It is also shown how a person 50 approaches the motor vehicle 1 from a first position 501 until they reach a second position 502, where they press the door opener button 42 in the door handle 41 and then step back to the side to a third position 503 in anticipation of the door 4 opening automatically. It should be noted that positions 502 and 503 are so close to the vehicle that the position 502, 503 of the person 50 cannot be trilaterated because the person 60 at positions 502 and 503 is only in the field of view of the ultrasonic sensor 14 installed in the door handle 41, but is outside the field of view of the next ultrasonic sensor 13 closest in the vehicle's longitudinal direction.

[0060] In step S1, the first unit 21 of the control device 2 measures the lateral surroundings 10 of the motor vehicle using the ultrasonic sensors 11-14 according to the measuring principles previously described, in particular with reference to Fig. 2 and Fig. 3, and generates a number of object features 701-703 that indicate the positions of suspected obstacles 50 in the surroundings 10 at the time of the measurement. Each generated object feature 701-703 is assigned a specific retention period during which the object feature 701-703 is retained and thus taken into account in the subsequent steps S2 and S3 before being discarded. The measurement is repeated at regular intervals to update the retained number of object features 701-703, so that the surroundings 10 of the motor vehicle 1 are continuously monitored.The retention time of a respective object feature 701 - 703 can also be selected to be longer than one measuring cycle in order to reduce "flickering" of the object features 701 - 703.

[0061] In step S2, the second unit 22 of the control device 2 checks whether a non-trilaterated object feature 702 has moved from a first area within a first distance yi from the door 4 to a second area between a second distance y2 and a third distance ya in front of the door. In Fig. 5, this is the case: When the door opener button 42 is pressed, the person 50 is at position 502 and the associated object feature 702 is within a distance yi from the door 4. After the door opener button 42 is pressed, the person 50 has stepped aside, but their new position 503, like position 502, cannot be correctly trilaterated, and the associated object feature 703 is therefore placed on the sensor axis of the ultrasonic sensor 14 in the second area between the distances y2 and ys from the door 4.Based on this movement pattern, the second unit 22 recognizes a typical movement pattern of a person 50 who has pressed the door opener button 42, stepped aside, and wishes the door 4 to be opened automatically. For this reason, the second unit 22 masks the object feature 703, which was incorrectly positioned within the pivoting range 40 of the door 4 due to failed trilateration, so that it is not taken into account in the subsequent step S3.

[0062] In step S3, the third unit 23 of the control device 21 determines, in response to the actuation of the door opener button 42, whether one or more of the object features 701-703 determined in steps S1 and not masked out in S2 are located within the pivoting range 40 of the door 4 to be opened. The third unit 31 then initiates an automatic opening of the door 4 depending on the object features 701-703 determined to be located within the associated pivoting range 40.

[0063] For example, the third unit 23 opens the door 4 only or only when there is no longer any object feature 701-703 in the pivoting range 40. Alternatively, the third unit 31 can determine a maximum possible opening angle α up to which the door 4 can be opened without touching one of the object features 703, and can only open the door 4 up to the determined maximum possible opening angle α. In this case, it is preferably taken into account that getting into the motor vehicle 1 is only comfortably possible from a minimum opening angle α0 of, for example, approximately 26°, and the door 4 can, for example, only be opened automatically if the determined maximum possible opening angle α0 is at least as large as the minimum opening angle α0. The maximum possible opening angle α shown in Fig. 5 is approximately 16°.Accordingly, the third unit 23 could not open the door 4 if the object feature 703 is located among the object features determined in step S1, even though the person 50 has actually stepped aside to a position 503 outside the pivoting range 40 of the door 4. However, since the incorrectly positioned object feature 703 was masked out as suggested in step S2 and the retention period of the previously determined object feature 702 has already expired, nothing stands in the way of opening the door 4, and the door 4 can advantageously be successfully opened automatically.

[0064] It should be noted that the person skilled in the art can easily select the first distance yi and the second and third distance y2 in the context of routine tests with a specific vehicle prototype and a series of test subjects who are instructed to press the door opener button 42 and step aside while observing the specific object features 701-703, such that the movement pattern of the test subjects can be reliably recorded and distinguished from other typical movements in the lateral environment 10 of the motor vehicle 1.

[0065] Fig. 6 illustrates the method according to a second embodiment using a plan view of a lateral environment 10 of the motor vehicle 1.

[0066] The second exemplary embodiment is based on the first exemplary embodiment. According to the second exemplary embodiment, a search window 8 is defined in the lateral surroundings 10 of the motor vehicle 1. The search window 8 extends in the vehicle's longitudinal direction from a point in front of the ultrasonic sensor 14 to a point behind a rear end of the pivoting range 40 of the door 4 in the vehicle's longitudinal direction, and extends in the vehicle's transverse direction from the door 4 or at least from a point within the first distance yi from the door to at least the third distance ya or even to the outer end of the pivoting range 40 in the vehicle's transverse direction. In other words, the search window 8 encompasses the first area within the distance yi and the second area between the second distance y2 and the third distance ya, but limits these areas to a predetermined section in the vehicle's longitudinal direction in the vicinity of the door opener button 42.

[0067] In step S2, according to the second embodiment, when determining whether an object feature 702 has moved from the first area within the first distance yi to the second area between the second distance y2 and the third distance ya, only those object features 702 that are located within the search window 8 are taken into account.

[0068] As a result, the movement pattern, the recognition of which triggers a masking of the object feature 730, is advantageously restricted even more precisely to the area around the door opener button 42, and similar movements that are carried out at other locations in the lateral environment 10 of the motor vehicle do not lead to an erroneous masking of object features that may be relevant for collision avoidance.

[0069] Fig. 7 illustrates process details of step S2 of masking an object feature 703 according to a third embodiment. The third embodiment is based on the second embodiment and also uses the search window 8. The third embodiment will now be described with reference to Fig. 1, Fig. 4, Fig. 6, and Fig. 7.

[0070] According to the third embodiment, steps S1, S2 and S3 are executed in parallel and cyclically.

[0071] The third exemplary embodiment describes in particular an advantageous embodiment of a sequence of step S2 (Fig. 4), with which, on the one hand, the movement pattern of the person 50 can be limited even more precisely in order to further reduce the erroneous masking of object features that cannot be assigned to the person 50 and the associated risk of collisions when opening the door 4, and with which, on the other hand, an advantageously particularly simple monitoring of the person 50 with little computational effort in a simple, sequential, cyclical sequence or loop is possible. The steps described below are substeps of the sequence of execution of step S2 shown in more detail in Fig. 7.

[0072] In step S201, the process of Fig. 7 begins. All timers and counters described below (in particular, the timers Ti and T2 described later and the validity counter C described later) are set to zero or reset, and the process proceeds to step S202.

[0073] In step S202, the system waits for the next measurement in step S1. This means that the system waits for an update of the stored object features 701, 702, 703. A check is then made to determine whether a first timer Ti has already exceeded a predetermined value ti, i.e., it checks whether T1 >t1 or whether a predetermined time ti has already elapsed. If no (N in S202), the system branches to step S203; if yes (Y in S202), the system branches to step S207.

[0074] In step S203, it is checked whether any object feature 8 is present in the search window 8. If no (N at S203), the process branches to step S204; if yes (Y at S203), the process branches to step S205.

[0075] In step S204, there is no object feature in the search window 8. Therefore, a validity counter C is incremented (increased by one). The validity counter C indicates how many cycles have been completed in which the search window 8 was in a defined state without any object features therein. A cycle can be understood as a cycle of steps S202, S203, S204 and / or steps S202, S207 (described below), S203, S204. A cycle can also be understood as a measurement cycle of step S1, since in each step S202, an update of the object features 701, 702, 703 is first awaited. After the validity counter C is incremented, the loop continues with step S202. If, however, an object feature 701, 703 is found within the search window 8 in step S203, the process branches to step S205 rather than to step S204.In step S205, the validity counter C is reset to zero because the sequence of valid cycles without object features is interrupted. The program then continues with step S206.

[0076] In step S206, the object feature 702, 703 identified in the search window 8 is transferred to the sequence of step S3, which is executed parallel to step S2. In other words, the object feature 702, 703 identified in the search window 8 is not hidden, even if it would otherwise meet all criteria for hiding, because the search window 8 has not been in a valid, defined state without object features long enough since the beginning of the sequence shown in Fig. 7. Therefore, at this point in time, the door 4 cannot be opened from step S3 if the object feature 702, 703 identified in the search window 8 is located within the pivoting range 40 of the door 4. Subsequently, the process jumps back to step S202.

[0077] If it is determined in step S202 that the predetermined time t1 has elapsed (T1 > t1 ;

[0078] Y at S202), the process branches to step S207.

[0079] In step S207, a check is made to determine whether the validity counter C has reached a predetermined value c, ie, whether C>c. If not (N at S207), the process branches back to step S203. If yes (Y at S207), the process branches to step S208.

[0080] Steps S202, S203, S204, S205, S206, S207 form a hysteresis loop that ensures that the program jumps to step S208 and considers masking out object features 703 in the manner described below only when, on the one hand, at least a predetermined time ti has elapsed since the start of the sequence in Fig. 2, and, on the other hand, a predetermined number c of measurement cycles has not detected any object feature in the search window 8. This hysteresis loop ensures that the search window 8 is in a valid, defined state without any other obstacles before the person 50 who wishes to press the door opener button 42 appears. In step S208, the second unit 22 of the control device 2 checks whether a non-trilaterated object feature 702, 703 is located in the search window 8. If no (N at S208), return to step S202.From there, since Ti>ti and C>c still hold, the system returns to step S208, and this loop is now run through until a non-trilaterated object feature 702, 703 is located in the search window 8. In this case (Y at S208), the system continues with step S209.

[0081] In step S209, a check is made to determine whether a dynamic object feature 702 is located in the search window 8. A dynamic object feature 702 is an object feature that has a speed whose magnitude exceeds a predefined threshold at the current time or has exceeded a predefined threshold at least once within a predefined time interval prior to the current time. In other words, a dynamic object feature 702 is an object feature that is moving or is known to have recently moved (so-called hysteresis). If no (N at S209), i.e., if there is no dynamic object feature in the search window 8, the process branches to step S206, i.e., a non-dynamic object feature 702 located in the search window 8 is not hidden in this phase, and the loop continues via steps S206, S202, S207, and S208.However, if a non-trilaterated and dynamic object feature 702 is detected in the search window 8 (Y at S209), the process continues with step S210.

[0082] It should be noted that in the position 502 of the person 50 shown in Fig. 6, the associated object feature 702 is dynamic, even if the person 50 briefly remains in front of the door opener button 42 while pressing the door opener button 42, since the person has previously moved from the position 501 into the first area up to the position 502. In step S210, it is checked whether the dynamic, non-trilaterated object feature 702 is located within the first distance yi from the door 4. If yes (Y at S210), this is an indication that the person is touching the door opener button 42. In this case, the process continues with step S213.

[0083] In step S213, a second timer T2 is set to zero. The second timer T2 is used to measure how much time has elapsed since the person 50 left the position 502 within the first distance yi from the door 4. The process then continues with step S206. This means that the object feature 702 continues to be hidden during this phase, and the loop continues via steps S206, S202, S207, S208, S209, and S210 until the person 50 moves out of the first area within the distance yi.

[0084] If it is then determined in step S210 that the dynamic, non-trilaterated object feature 702 is no longer within the first distance yi to the door 4 (N at S210), the process branches to step S211.

[0085] In step S211, a check is made to determine whether the second timer T2 is greater than a predetermined value t2, T2>t2. If yes (Y at S211), more than the predetermined time t2 has already elapsed since the person 50 moved away from the door opener button 42. The program branches to step S206, meaning that the object feature 702 continues to be hidden, and the loop continues. According to a modification, the process can also be terminated at this point and restarted with step S201.

[0086] However, if it is determined in step S211 that the predetermined time t2 has not yet elapsed (N at S21 1 ), the process proceeds to step S212.

[0087] In step S212, it is checked whether a non-trilaterated object material 703 is now located in the second area between the distances y2 and ya to the door within the search window 8. At this point, it is not required that the object feature 703 must be a dynamic object feature 703, since the person 50 can remain relatively motionless in position 503 in anticipation of the automatic opening of the door 4.

[0088] If no non-trilaterated object feature 703 is found in the second area (N at S212), the process branches to step S206, i.e., any non-trilaterated object feature present elsewhere is taken into account in step S3 and not hidden, and the loop continues via steps S206, S202, S207, S208, S209, S210 and S211.

[0089] However, if it is determined in step S212 that a non-trilaterated object feature 703 is located in the second area (Y at S212), the process branches to step S214.

[0090] In this case, the process assumes that the non-trilaterated object feature 703 is mistakenly located within the search window 8 or within the pivoting range 40 of the door 4 due to the circumstances previously explained with reference to Figs. 3 and 5, and that the person 50 has actually stepped out of the pivoting range 40 of the door. Therefore, the non-trilaterated object feature 703 is hidden in step S214.

[0091] The loop then continues through steps S202 to S211 until the predetermined time t2 has elapsed in step S211. From this point on, the non-trilaterated object feature 703 is no longer hidden.

[0092] According to the third embodiment, step S3 is executed parallel to or independently of step S2 and the sequence shown in Fig. 7. Step S3 can be executed when the person 50 presses the door opener button 42. The third unit 23 then checks whether a non-masked object feature 701, 702, 703 is present in the search window 8 and, more specifically, in the pivoting range 40 of the door. If so, the third unit 23 waits until the timer T2 expires. If the object feature 703 is masked within this time, and there are then no further object features 701, 702 in the pivoting range 40 of the door 4, the third unit 23 initiates the automatic opening of the door 4.If, on the other hand, at least one non-masked object feature 703 is still present in the pivoting range 40 of the door 4 when the timer T2 expires, the third unit 23 either does not open the door 4, or it determines the maximum opening angle a up to which the door 4 can be opened without touching any of the non-masked object features 703 in the pivoting range 40, and opens the door 4 only up to the opening angle a - but only if the determined opening angle a is above a minimum value of, for example, 26°.

[0093] After the door 4 is closed again and / or after the timer T2 has expired without the door 4 being able to be opened, the third unit 23 causes at least the timer T2 and the validity counter c of the second unit 22 to be reset and / or that the process from Fig. 7 is completely restarted with step S201.

[0094] According to the third embodiment, a technique has been described with which the movement pattern of person 50 can be advantageously captured even more precisely in a simple, sequentially executed loop. According to the third embodiment, the object feature 703 erroneously placed in the pivoting area 40 of door 4 is only masked out if the following conditions are met:

[0095] Condition 1: Before the first appearance of the object feature 702 in the first area within the distance yi to the door, a predetermined time ti must have elapsed, and the search window 8 must have been free of object features for a predetermined number of passes through the loop S202-S207 or a predetermined number of ultrasonic measurements.

[0096] Condition 2: The object feature 702 was recognized as a dynamic and non-trilaterated object feature 702 during its passage through the first area and was located within the search window 8. Condition 3: Since the object feature 702 left the first area within the first distance, less than a predetermined time t2 has elapsed.

[0097] Condition 4: The object feature 703 to be hidden is not trilaterated and is located within the second area between the second distance y2 and the third distance y3 in front of the door and within the search window 8.

[0098] Accordingly, a typical movement pattern of a person 50 wishing to open the door 4 can advantageously be recognized particularly precisely, and only if this movement pattern is present can the object feature 703 incorrectly placed in the pivoting area 40 be masked out. However, other movement patterns do not result in the object feature 703 being masked out. Thus, an optimal balance is achieved between the desire to protect the door 4 from collisions during automatic opening and the desire to quickly and correctly open the door 4 with an obstacle-free pivoting area 40.

[0099] The described exemplary embodiments can be used particularly advantageously in particular when a manufacturer's specification exists to the effect that, for example, ultrasonic sensors 13, 14 are to be installed only in the door handles 31, 41, either no ultrasonic sensors are to be installed in the side sill 5, or the ultrasonic sensors are to be installed in the side sill 5 in the vehicle's longitudinal direction at substantially the same position as the ultrasonic sensors 13, 14 in the door handles 31, 41. In both cases, non-trilaterable object features are to be expected to be increasingly present, which can be masked out according to the teachings of the described exemplary embodiments.

[0100] Although the present invention has been described using exemplary embodiments, it is susceptible to numerous modifications. The hysteresis loop from steps S202-S207 described in the third exemplary embodiment is optional and can also be omitted, i.e., the sequence can also jump directly to step S208 after step S201, step S206, or step S214.

[0101] The automatic opening of the left rear door 4 of the motor vehicle 1 was described. However, the teachings described are also applicable to the automatic opening of the right rear door and / or the left or right front door 3 of the motor vehicle 1 and develop the same advantages here, in particular if no further ultrasonic sensors are installed in the side sill 5 of the motor vehicle 1. It is also conceivable to apply the teachings described to a rear door of the motor vehicle if, for example, an ultrasonic sensor is installed only in one door handle of the rear door. It is also conceivable to apply the proposed method to motor vehicles that have only a single door on each side or more than two doors on each side.

[0102] The embodiments were described using point-shaped object features 71, 72, 701-703. However, the described teaching is also applicable to differently shaped object features, for example, to linear object features, as long as they are constructed from non-trilaterizable individual measurements.

[0103] The positions at which ultrasonic sensors 11, 12, 13, 14 are arranged in Fig. 1 are purely exemplary. In particular, the ultrasonic sensors 11, 12 are entirely optional. The ultrasonic sensors 13, 14 also do not necessarily need to be installed in the door handles 31, 41; the described advantages of the method are also achieved if the ultrasonic sensors 13, 14 are installed at a different location on the respective door or, for example, not on the door but in the area of ​​the respective door in the side sill 5.To achieve the technical advantages of the proposed method, it is only necessary that the ultrasonic sensor 14 is arranged in such a way that it sees a person 50 who operates the door opener button 42 (the person 50 is located within a signal beam emitted by the ultrasonic sensor 14), while one or more further ultrasonic sensors 13 that are sufficiently far apart from the ultrasonic sensor 14 in the longitudinal direction of the vehicle for trilateration do not see this person 50 if they are standing directly in front of the door opener button 42.

[0104] The proposed method is therefore particularly advantageous if only one ultrasonic sensor 13, 14 is to be installed on one side of the motor vehicle 1 in the area of ​​a respective door 3, 4, for example, if ultrasonic sensors 13, 14 are to be installed only in the door handles 31, 41, and no other ultrasonic sensors are to be installed to monitor the lateral environment 10 of the motor vehicle 1. Even with such strict manufacturer specifications, the proposed method can reliably open the door 4 after the person 50 has actuated the door opener button 42.

[0105] However, this is not a necessary limitation, and the method also offers advantages when multiple ultrasonic sensors 13, 12 are installed in the area of ​​a door 3, but are only a short distance apart in the vehicle's longitudinal direction. Even in such cases, trilateration of an object feature 702 of a person 50 located close to the door 3 may fail in certain cases; accordingly, the proposed method can also be advantageously applied here.

[0106] LIST OF REFERENCE SYMBOLS

[0107] 1 motor vehicle

[0108] 2 Control device

[0109] 21-23 units of the control device

[0110] 3 front door

[0111] 4 rear door

[0112] 5 side skirts

[0113] 6 front bumper

[0114] 8 search windows

[0115] 10 Lateral surroundings of the motor vehicle

[0116] 11-14 Ultrasonic sensors

[0117] 31 Door handle

[0118] 32 door opener button

[0119] 41 Door handle

[0120] 42 Door opener button

[0121] 44 Pivot point of the front door

[0122] 50 Obstacle, Person

[0123] 51 , 52 obstacle

[0124] 62-64 Locus curves of possible reflection points

[0125] 71 , 72 Object features

[0126] 501 first position

[0127] 502 second position

[0128] 503 third position

[0129] 701 Object feature

[0130] 702, 703 non-trilateable object features

[0131] S1-S3 process steps

[0132] S201 -S213 steps of a loop that implements process step S2 y1-y3 distances in the vehicle transverse direction

Claims

PATENT CLAIMS 1. A method for determining obstacles (50) in a pivoting range (40) of a door (4) of a motor vehicle (1) by means of a number of ultrasonic sensors (13, 14) mounted on the motor vehicle (1), comprising: a) determining (S1) a number of object features (701-703), each indicating a position of a suspected obstacle (50) in the surroundings (10) of the motor vehicle (1), by means of distance measurements using a number of ultrasonic sensors (13, 14) of the motor vehicle (1), wherein the position (501) of an obstacle (50) measured by more than one of the ultrasonic sensors (13, 14) is determined by means of trilateration using the respective measured distances and the positions of the respective ultrasonic sensors (13, 14), and the respective object feature (701) is marked as trilaterated, and the position (502, 503) of an obstacle (50), which was measured by only one of the ultrasonic sensors (14),is determined according to the measured distance and a field of view of the respective ultrasonic sensor (14) and the respective object feature (702, 703) is marked as non-trilaterated, and b) masking (S2) a non-trilaterated object feature (703) that has moved from a first area within a first distance (yi) to the door (4) to a second area between a second (y2) and third distance (y3) to the door (4), wherein the first distance (yi) is smaller than the second distance (y2) and the second distance (y2) is smaller than the third distance (y3).

2. Method according to claim 1, characterized in that Steps a) and b) are carried out cyclically and in step b) the object feature (703) is only hidden if a predetermined time (ti) has elapsed before the object feature (702) appears in the first area and no object features have been determined at least in the first area and in the second area over a predetermined number of runs of step a).

3. Method according to one of the preceding claims, characterized in that Step b) is carried out only for non-trilaterated object features (703) in the second region if a non-trilaterated object feature (702) is previously detected in the first region, which has a speed whose magnitude exceeds a predetermined threshold value.

4. Method according to one of the preceding claims, characterized in that in step b) the object feature (703) is only masked out if less than a predetermined time has elapsed since the object feature (702) left the first area within the first distance (yi).

5. Method according to one of the preceding claims, characterized in that in step b) only object features (702, 703) are treated which are located within a predefined search window (8) which extends in the vehicle longitudinal direction from a position in the vehicle longitudinal direction in front of a position of a door handle (41) of a door (4) of the motor vehicle (1) to a position at the rear end of the pivoting range (40) of the door (4) in the vehicle longitudinal direction and extends in the vehicle transverse direction from in front of the first distance (yi) to the motor vehicle (1) to at least the third distance (y3) to the motor vehicle.

6. Method according to one of the preceding claims, further comprising: c) automatically opening (S3) the door (4) of the motor vehicle (1) in response to actuation of a door opener button (42) attached to the door (4) as a function of the determined and non-masked object features (702, 703) which are located within the pivoting range (40) of the door (4).

7. Method according to claim 6, characterized in that Step c) comprises opening the door (4) only if there are no object features in the pivoting range (40) of the door (4), or in step c) a maximum opening angle (a) for the door (4) is determined based on the object features (703) located within the pivoting range (40) of the door (4), and the door (4) is only opened up to the specific maximum opening angle (a) and this only if the specific maximum opening angle (a) is greater than a specified minimum opening angle.

8. A computer program product comprising instructions which, when the program is executed by a computer (2), cause the computer (2) to carry out the method according to one of claims 1 to 7.

9. Control device (2) for a motor vehicle (1) for determining obstacles (50) in a pivoting range (40) of a door (4) of the motor vehicle (1) by means of a number of ultrasonic sensors (13, 14) attached to the motor vehicle (1), comprising: a) a first unit (21) which is designed to determine a number of object features (701-703), each of which indicates a position of a suspected obstacle (50) in the environment (10) of the motor vehicle (1), by means of distance measurements using a number of ultrasonic sensors (13, 14) of the motor vehicle (1), wherein the position (501) of an obstacle (50) measured by more than one of the ultrasonic sensors (13, 14) is determined by means of trilateration using the respective measured distances and the positions of the respective ultrasonic sensors (13, 14), and the respective object feature (701) is used as trilaterated, and the position (502,530) of an obstacle (50) measured by only one of the ultrasonic sensors (13, 14) is determined according to the measured distance and a field of view of the respective ultrasonic sensor (14), and the respective object feature (702, 703) is identified as non-trilaterated, and b) a second unit (22) configured to mask out a non-trilaterated object feature (703) that has moved from an area within a first distance (yi) to the door (4) to an area between a second (y2) and third distance (y3) to the door, wherein the first distance (yi) is smaller than the second distance (y2) and the second distance (y2) is smaller than the third distance (y3).

10. Motor vehicle (1) with a control device (2) according to claim 9.

Images