PatentDe  


Dokumentenidentifikation EP1602940 22.11.2007
EP-Veröffentlichungsnummer 0001602940
Titel Vorrichtung und Verfahren zu Hinderniserkennung mit einem Radarsensor und einer Kamera
Anmelder Toyota Jidosha Kabushiki Kaisha, Toyota, Aichi, JP
Erfinder Tokoro, Setsuo, Toyota-shi, Aichi 471-8571, JP
Vertreter WINTER, BRANDL, FÜRNISS, HÜBNER, RÖSS, KAISER, POLTE, Partnerschaft, 85354 Freising
DE-Aktenzeichen 602005002783
Vertragsstaaten DE, FR, GB
Sprache des Dokument EN
EP-Anmeldetag 01.06.2005
EP-Aktenzeichen 050118512
EP-Offenlegungsdatum 07.12.2005
EP date of grant 10.10.2007
Veröffentlichungstag im Patentblatt 22.11.2007
IPC-Hauptklasse G01S 13/86(2006.01)A, F, I, 20070911, B, H, EP
IPC-Nebenklasse G01S 7/40(2006.01)A, L, I, 20070911, B, H, EP   G01S 13/93(2006.01)A, L, I, 20070911, B, H, EP   G05D 1/02(2006.01)A, L, I, 20070911, B, H, EP   

Beschreibung[en]
BACKGROUND OF THE INVENTION Field of the Invention

The present invention relates to an obstacle recognition system and an obstacle recognition method for effecting recognition of an obstacle to vehicle travel.

Related Background of the Invention

An example of the conventionally known obstacle recognition systems is one having a millimeter wave radar and an image acquiring means, as described in Japanese Patent Application Laid-Open No. 2001-296357 . This system is configured to specify an image recognition area on the basis of the power outputted from the millimeter wave radar, to execute image processing for detection of an object within the limited image recognition area thus specified, and thereby to reduce the time of the image processing.

US 2004/0080449 A1 describes a radar device having a camera and a laser radar. A control unit corrects the direction of the center axis of the laser radar on the basis of the image obtained by the camera.

In US 6,580,385 B1 a object detection system is described which is made up of a combination of three different object detectors, each of which being of a different type and having a different detection area in front of the vehicle. Detection data being obtained in overlapping zones of the different object detectors are used for other analyses such as object tracking and plausibility check of the detector data.

SUMMARY OF THE INVENTION

However, these systems have a problem that appropriate obstacle recognition is not performed. For example, where there exists an object with low reflection of waves, the millimeter wave radar fails to perform accurate recognition of the object. In a reverse case where there exists an object with large reflection of waves, the object, such as a metal bump on a road, poses no serious impediment to vehicle travel in some situations, and it makes appropriate recognition of an obstacle difficult.

An object of the present invention is therefore to provide an obstacle recognition system and an obstacle recognition method capable of performing appropriate recognition of an obstacle.

Namely, an obstacle recognition system according to the present invention comprises a radar for detecting existence of an object around a vehicle, image taking means for taking an image of the object, obstacle detecting means for detecting an obstacle to travel of the vehicle on the basis of an output from the radar, and criterion value changing means for changing a criterion value in the obstacle detection of the obstacle detecting means on the basis of image information of the image taken by the image taking means.

The obstacle recognition system according to the present invention is characterized in that the system further comprises obstacle estimating means for estimating presence of the obstacle on the basis of the image information and in that the criterion value changing means changes the criterion value on the basis of a possibility of the presence of the obstacle estimated by the obstacle estimating means.

The obstacle recognition system according to the present invention is characterized in that the obstacle estimating means sets a detection area in the image on the basis of the output from the radar, and estimates the presence of the obstacle in the detection area on the basis of the image information.

The obstacle recognition system according to the present invention is characterized in that the detection area set by the obstacle estimating means is set based on a distance from the vehicle to the object and a relative position of the object to the vehicle.

The obstacle recognition system according to the present invention is characterized in that the obstacle estimating means estimates the presence of the obstacle on the basis of an edge density distribution obtained through image processing of the image.

The obstacle recognition system according to the present invention is characterized in that the obstacle estimating means calculates a possibility of absence of the obstacle on the basis of entire edge data obtained through image processing of a predetermined region in the image, calculates the possibility of the presence of the obstacle on the basis of edge data resulting from subtraction of edge data of lane markers from the entire edge data, and estimates the presence of the obstacle on the basis of the possibility of the absence of the obstacle and the possibility of the presence of the obstacle.

The obstacle recognition system according to the present invention is characterized in that the criterion value changing means changes the criterion value so that the obstacle becomes less likely to be detected as the possibility of the absence of the obstacle is estimated to be higher based on the image information.

The obstacle recognition system according to the present invention is characterized in that the criterion value changing means changes the criterion value so that the obstacle becomes more likely to be detected as the possibility of the presence of the obstacle is estimated to be higher based on the image information.

An obstacle recognition method according to the present invention comprises a criterion value changing step of changing a criterion value in obstacle detection on the basis of image information of an image taken by an image taking device for taking an image around a vehicle; and an obstacle detecting step of detecting an obstacle to travel of the vehicle on the basis of an output from a radar for detecting existence of an object by use of the criterion value set in the criterion value changing step.

The obstacle recognition method according to the present invention is characterized in that the method further comprises an obstacle estimating step of estimating presence of the obstacle on the basis of the image information and in that the criterion value changing step comprises changing the criterion value on the basis of a possibility of the presence of the obstacle estimated in the obstacle estimating step.

The obstacle recognition method according to the present invention is characterized in that the obstacle estimating step comprises setting a detection area in the image on the basis of the output from the radar and estimating the presence of the obstacle in the detection area on the basis of the image information.

The obstacle recognition method according to the present invention is characterized in that the detection area in the obstacle estimating step is set based on a distance from the vehicle to the object and a relative position of the object to the vehicle.

The obstacle recognition method according to the present invention is characterized in that the obstacle estimating step comprises estimating the presence of the obstacle on the basis of an edge density distribution obtained through image processing of the image.

The obstacle recognition method according to the present invention is characterized in that the obstacle estimating step comprises calculating a possibility of absence of the obstacle on the basis of entire edge data obtained through image processing of a predetermined region in the image, calculating a possibility of presence of the obstacle on the basis of edge data resulting from subtraction of edge data of lane markers from the entire edge data, and estimating the presence of the obstacle on the basis of the possibility of absence of the obstacle and the possibility of presence of the obstacle.

The obstacle recognition method according to the present invention is characterized in that the criterion value changing step comprises changing the criterion value so that the obstacle becomes less likely to be detected as the possibility of absence of the obstacle is estimated to be higher based on the image information.

The obstacle recognition method according to the present invention is characterized in that the criterion value changing step comprises changing the criterion value so that the obstacle becomes more likely to be detected as the possibility of presence of the obstacle is estimated to be higher based on the image information.

According to these aspects of the invention, the criterion value in the obstacle detection is changed based on the image information and it is feasible thereby to implement the following change of the criterion value: in the obstacle detection based on the output from the radar, the criterion value is changed so that the obstacle becomes less likely to be detected as the possibility of absence of the obstacle becomes higher based on the image information; the criterion value is changed so that the obstacle becomes more likely to be detected as the possibility of presence of the obstacle becomes higher. This can prevent unwanted actuation of the system when the possibility of presence of the obstacle is low. The system can be actuated more securely when the possibility of presence of the obstacle is high. This achieves appropriate actuation of the system.

The obstacle recognition system according to the present invention is also characterized in that it further comprises collision avoiding/relieving means for avoiding a collision with the obstacle or for relieving influence of the collision and in that the collision avoiding/relieving means activates the collision avoidance or collision influence relief earlier when the obstacle estimating means estimates that the possibility of the presence of the obstacle is high, than when the possibility of the presence of the obstacle is estimated not to be high.

The obstacle recognition method according to the present invention is also characterized in that it further comprises a collision avoiding/relieving step of avoiding a collision with the obstacle or relieving influence of the collision and in that the collision avoidance or collision influence relief in the collision avoiding/relieving step is activated earlier when the obstacle estimating step results in estimating that the possibility of the presence of the obstacle is high, than when the possibility of the presence of the obstacle is estimated not to be high.

According to this invention, the collision avoidance or collision influence relief with the obstacle is activated earlier with the possibility of presence of the obstacle being high, whereby it is feasible to enhance the possibility of collision avoidance with the obstacle or to relieve influence of the collision.

[Effect of the Invention]

The present invention involves the execution of the obstacle detection with the radar on the basis of the image information, and thus implements appropriate recognition of the obstacle.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is a schematic diagram of a configuration of an obstacle recognition system according to an embodiment of the present invention.

  • Fig. 2 is a flowchart showing an operation of the obstacle recognition system of Fig. 1 and an obstacle recognition method.
  • Fig. 3 is an illustration showing a photographic image taken by an image taking part 2 in the obstacle recognition system of Fig. 1.
  • Fig. 4 is an illustration to illustrate a control processing content in the flowchart of Fig. 2.
  • Fig. 5 is an illustration to illustrate a control processing content in the flowchart of Fig. 2.
  • Fig. 6 is an illustration to illustrate a control processing content in the flowchart of Fig. 2.
  • Fig. 7 is an illustration to illustrate a control processing content in the flowchart of Fig. 2.

DESCRIPTION OF THE PREFERRED ENBODIMENTS

An embodiment of the present invention will be described below in detail with reference to the accompanying drawings. The same elements will be denoted by the same reference symbols in the description of the drawings, without redundant description.

Fig. 1 is a schematic view of a configuration of an obstacle recognition system according to an embodiment of the present invention.

As shown in Fig. 1, the obstacle recognition system 1 of the present embodiment is a device mounted on a vehicle 5, and device for detecting an obstacle in front of the vehicle 5. This obstacle recognition system 1 is provided with an image taking part 2. The image taking part 2 functions as an image taking means for taking an image of an object around the vehicle, and is installed, for example, so as to take a forward image ahead the vehicle 5. This image taking part 2 is, for example, a CCD camera or a C-MOS camera.

The obstacle recognition system 1 has a radar 3. The radar 3 is a device for detecting the existence of an object around the vehicle, and is installed, for example, so as to be able to detect the existence of an object ahead the vehicle 5. The radar 3 is one configured to transmit an electromagnetic wave toward a detection area, to receive a reflected wave thereof, and to detect an object in the detection area, and is, for example, a millimeter wave radar that transmits and receives a millimeter wave. The radar 3 can be one configured to transmit and receive a wave other than electromagnetic waves and to detect an object, based thereon, and can be, for example, a laser radar for transmitting and receiving a laser beam, an acoustic wave radar for transmitting and receiving an acoustic wave, or the like.

The image taking part 2 and radar 3 are connected to an ECU (Electronic Control Unit) 4. The ECU 4 performs control over the entire system, and is comprised, for example, of a CPU, a ROM, a RAM, an input signal circuit, an output signal circuit, a power supply circuit, and so on. This ECU 4 functions as an obstacle detecting means for retrieving image information from the image taking part 2 and radar information from the radar 3 and for detecting an obstacle to vehicle travel. The ECU 4 also functions as a criterion adjusting means for adjusting a criterion value in obstacle detection processing on the basis of the image information from the image taking part 2. The ECU 4 also functions as an obstacle estimating means for estimating the presence of an obstacle on the basis of the image information.

This obstacle recognition system 1 is preferably used as a part of a vehicle driving support system for avoidance of a collision of a vehicle. This vehicle driving support system executes a warning process for issuing a warning to a driver of a vehicle, a brake preparation process for collision avoidance, and part or all of brake control for collision avoidance, based on obstacle detection information detected by the obstacle recognition system 1. By executing these processes, it is feasible to avoid a collision with an obstacle or to relieve influence of the collision with the obstacle.

The obstacle recognition system 1 is preferably installed on a vehicle in combination with the vehicle driving support system for lane keep. In this configuration, the lane keep control can be performed using the image information on lane markers among the image information obtained by the image taking part 2, and the obstacle detection can be performed using the image information except for the lane markers, thereby utilizing the image information without waste. Namely, the image information except for the lane markers, which was discarded in the lane keep control, can be effectively used, so as to implement efficient obstacle detection processing.

The operation of the obstacle recognition system and an obstacle recognition method according to the present embodiment will be described below.

Fig. 2 is a flowchart showing the operation of the obstacle recognition system and the obstacle recognition method according to the present embodiment. Figs. 3 to 6 are illustrations to illustrate a photographic image, arithmetic processing, etc. in the control processing of Fig. 2. The control processing in Fig. 2 is repeatedly executed in given cycles preliminarily set by the ECU 4.

First, the radar information is retrieved as at S10 in Fig. 2. This retrieval of the radar information is carried out by retrieving detected object information based on a detection signal of the radar 3. The object information based on the detection signal of the radar 3 is, for example, information about a relative distance between the vehicle and an object, a relative speed, and a direction of the object. The radar 3 performs a process of acquiring the object information about these items from the detection signal of the radar 3. The ECU 4 may be configured to perform the process of acquiring the object information about these items. For example, the ECU 4 may be arranged to receive the detection signal of the radar 3 and to perform an arithmetic operation to acquire the information about the relative distance between the vehicle and the object, the relative speed, and the direction of the object on the basis of the detection signal.

Then the ECU 4 moves to S12 to retrieve the image information. This retrieval of the image information is carried out by retrieving object information detected based on the photographic image of the image taking part 2. The object information based on the photographic image of the image taking part 2 is, for example, vertical edge information obtained through edge processing of the photographic image. Specifically, it is edge density distribution information in the vertical direction within a predetermined region of the photographic image. A process of acquiring this vertical edge information is carried out by the image taking part 2. The ECU 4 may be configured to perform this process of acquiring the vertical edge information. For example, the ECU 4 may be arranged to receive the photographic image of the image taking part 2 and to perform the image processing of the photographic image to acquire the vertical edge information.

Then the ECU 4 moves to S14 to determine whether a determination on an obstacle based on image information is available. In this availableness determining process, it is determined that the obstacle determination based on the image information is not available, for example, where it is determined that the image information is not available, where a fail signal is received from the image taking part 2, where an error occurs in communication with the image taking part 2, where the vehicle speed is not more than a predetermined speed, and so on. The ECU 4 terminates the control processing when the obstacle determination is determined to be unavailable as described.

On the other hand, when it is determined at S14 that the obstacle determination is available, the ECU 4 performs an obstacle estimating process based on the image information (S16). This obstacle estimating process is a process of estimating a possibility of presence of an obstacle on a road on the basis of the image information; for example, the ECU 4 performs an obstacle absence flag operation and an obstacle presence flag operation, and estimates the possibility of the presence of the obstacle on the basis of the presence/absence of these obstacle absence flag and obstacle presence flag, or the number of flags set, or the like.

The obstacle absence flag operation is a process of computing obstacle absence flags on the basis of edge points obtained through vertical edge processing for the photographic image of the image taking part 2. An obstacle absence flag is a flag indicating a possibility of absence of an obstacle on a road on the basis of the image information. The more the obstacle absence flags are set, the higher the possibility of the absence of the obstacle is.

For example, as shown in Fig. 3, a predetermined image processing region 31 is set in the photographic image 30 of the image taking part 2, and the vertical edge processing is carried out in the whole of this image processing region 31. Then, as shown in Fig. 4, a vertical edge density distribution is created based on edge points obtained by the edge processing. This edge density distribution is a histogram in which edge points are arranged in the lateral direction. In this Fig. 4, the edge points in the histogram include all edge points of lane markers 32, a part of vehicle 33, etc. in the photographic image 30. The lateral direction of this edge density distribution in Fig. 4 corresponds to the lateral direction of the photographic image 30 in Fig. 3. The vehicle 33 is another car traveling ahead the vehicle 5 being a host vehicle, and can be an obstacle to travel of the host vehicle.

A predetermined threshold 41 is set in the edge density distribution of Fig. 4. When a histogram level 42 at an edge point is below the threshold 41, an obstacle absence flag 43 is set. On that occasion, as a detection area 44 for obstacle absence flag 43, it is preferable to set a region where a vehicle can be estimated to exist based on the output of the radar 3, instead of the region corresponding to the whole of the image processing region 31. By limiting the detection area 44 in this manner, it is feasible to reduce the processing amount of the arithmetic operation and to shorten the processing time of the arithmetic operation. This detection area 44 is preferably set based on the distance from the vehicle 5 to the object (other car) and the relative position of the object to the vehicle 5.

This detection area 44 is divided into a plurality of subregions (regions separated by vertical dashed lines in Fig. 4). For example, twelve subregions are set in the detection area 44. It is then determined for each of the subregions whether a histogram level 42 exceeds the threshold 41, and an obstacle absence flag 43 is set in each subregion where the histogram level 42 is below the threshold 41.

On the other hand, the obstacle presence flag operation is a process of computing obstacle presence flags on the basis of the edge points obtained through the vertical edge processing of the photographic image of the image taking part 2. An obstacle presence flag is a flag indicating a possibility of presence of an obstacle on a road on the basis of the image information. The more the obstacle presence flags are set, the higher the possibility of the presence of the obstacle is.

For example, edge points concerning lane markers 32 are removed from the edge points obtained by the aforementioned obstacle absence flag operation, to extract obstacle edge points. This edge point removing process about lane markers 32 may be carried out by removing all edge points as a pair of a rise edge and a fall edge from the edge points obtained by the obstacle absence flag operation.

Then, as shown in Fig. 5, an edge density distribution is created as to the obstacle edge points. This edge density distribution is a histogram in which the obstacle edge points are arranged in the lateral direction. The lateral direction of this edge density distribution in Fig. 5 corresponds to the lateral direction of the photographic image 30 in Fig. 3.

A predetermined threshold 45 is set in the edge density distribution of Fig. 5. When a histogram level 46 at an obstacle edge point exceeds the threshold 45, an obstacle presence flag 47 is set. On that occasion, as a detection area 48 for obstacle presence flag 47, it is preferable to set a region where a vehicle can be estimated to exist based on the output of the radar 3, instead of the region corresponding to the whole of the image processing region 31. By limiting the detection area 48 in this manner, it is feasible to reduce the processing amount of the arithmetic operation and to shorten the processing time of the arithmetic operation.

For example, a plurality of subregions are set in this detection area 48. It is then determined for each of the subregions whether the histogram level 46 exceeds the threshold 45, and an obstacle presence flag 47 is set in each subregion where the histogram level 46 exceeds the threshold 45.

Then the ECU 4 moves to S18 in Fig. 2 to determine whether an obstacle absence flag is set. This determination is made based on whether the obstacle absence flag 43 is set in all the subregions in Fig. 4, in the image edge arithmetic operation at S16. Since in Fig. 4 the obstacle absence flag 43 is not set in all the subregions, the determination process at S18 results in determining that the obstacle absence flag is not set When the obstacle absence flag 43 is set in all the subregions on the other hand, the determination process at S 18 results in determining that the obstacle absence flag is set.

When the obstacle absence flag is set at S 18, it is determined that the possibility of the absence of the obstacle is high, and a threshold increase process is carried out (S24). This threshold increase process is a process of setting the determination threshold (criterion value) for determining the presence/absence of the obstacle, at a higher value so as to make an object on a road less likely to be determined as an obstacle in the obstacle detection process of detecting the obstacle to travel of the vehicle on the basis of the output of the radar 3.

For example, in a case where a determination threshold is set for the output of the radar 3 and where it is determined that an obstacle is present when the output exceeds the determination threshold, the determination threshold is switched to a higher threshold value to make an object on a road less likely to be determined as an obstacle.

When the obstacle absence flag is not set at S18, it is determined that the possibility of the absence of the obstacle is not low, and it is determined whether the obstacle presence flag is set (S20). This determination is made, for example, based on whether the obstacle presence flag 47 in Fig. 5 is set in a predetermined number of subregions or more, in the image edge arithmetic operation at S16. The predetermined number herein is preferably set to 2. In this case, since the obstacle presence flag 47 is set in two subregions in Fig. 5, it is determined in the determination process at S20 that the obstacle presence flag is set. On the other hand, when the obstacle presence flag 47 is not set in two or more subregions, it is determined in the determination process at S20 that the obstacle presence flag is not set.

When the obstacle presence flag is not set at S20, it is determined that the possibility of the presence of the obstacle is low, and the control processing is terminated. When the obstacle presence flag is set at S20 on the other hand, it is determined that the possibility of the presence of the obstacle is high, and a threshold decrease process is carried out (S22). This threshold decrease process is a process of setting the determination threshold (criterion value) for determining the presence/absence of the obstacle, at a lower value so as to make an object on a road more likely to be determined as an obstacle, in the obstacle detection process of detecting the obstacle to travel of the vehicle on the basis of the output of the radar 3.

For example, in a case where a high threshold is set for the output of the radar 3, the determination threshold is switched to a lower threshold value. This makes an object on a road more likely to be determined as an obstacle.

With a specific example, as shown in Fig. 6, there is a curve ahead a road where the host vehicle 5 is traveling, and another vehicle 33 is traveling on the curve; there is a case where it is desired to determine whether the other vehicle 33 is on the same lane as the host vehicle. In this case, as shown in Fig. 7, a probability of making a determination that the other vehicle 33 is on the same lane as the host vehicle is low at first, but the probability increases as the host vehicle approaches the other vehicle 33. At this time, a high criterion value 62 is set for the output of the radar 3. Since it is difficult to detect the width of the forward obstacle with the radar 3, the determination is made in a secure state in order to avoid unwanted actuation of the system. On the other hand, in a case where two edges are present ahead on the basis of the image information, where the size is estimated from the width thereof, and where it is anticipated that a vehicle or the like is present, the high criterion value 62 is changed to a low criterion value 63. This makes an object on a road more likely to be determined as an obstacle.

Then the ECU 4 moves to S26 in Fig. 2 to perform an obstacle detection process. This obstacle detection process is a process of detecting an obstacle to travel of the vehicle on the basis of the output of the radar 3. For example, as shown in Fig. 7, whether an obstacle is present is determined based on whether the output 61 of the radar 3 exceeds the determination threshold 62 or 63 set at S22 or 24. Namely, when the low determination threshold 62 is set as the determination threshold and when the output 61 of the radar 3 exceeds the determination threshold 62, it is determined that an obstacle is present When the output 61 of the radar 3 is below the determination threshold 62 on the other hand, it is determined that no obstacle is present.

Then the ECU 4 moves to S28 to perform a collision avoiding/relieving process. The collision avoiding/relieving process is a process of avoiding a collision of the vehicle 5 with the obstacle or relieving influence of the collision. Namely, in this collision avoiding/relieving process, the collision avoidance or collision influence relief is activated earlier when the presence of the obstacle is detected, than when the absence of the obstacle is detected. For example, when the presence of the obstacle is detected, a control system is initiated earlier in brake control of effecting automatic braking for collision avoidance or in steering control of effecting automatic steering for collision avoidance. In addition, when the presence of the obstacle is detected, a control system is initiated earlier in seat belt control of automatically retracting the seat belts for relieving the influence of the collision, in seat control of automatically moving the seats, in headrest control of automatically moving the headrests, or in suspension control of automatically changing a suspension characteristic.

A warning process of issuing a warning on the presence of the obstacle to the driver of the vehicle 5 may be carried out as the collision avoiding/relieving process at S28. It is preferable to issue a warning on the presence of the obstacle ahead the vehicle through visual sensation, auditory sensation, or tactual sensation to the driver, for example, by turning on a lamp, by outputting a sound, or the like, prior to the initiation of the automatic braking or the like described above.

As described above, the obstacle recognition system and obstacle recognition method according to the present embodiment are arranged to change the determination threshold in the obstacle detection on the basis of the image information, so that they can set the criterion value at the high value with the possibility of the presence of the obstacle being low or set the criterion value at the low value with the possibility of the presence of the obstacle being high, on the basis of the image information in the obstacle detection based on the output of the radar 3. This makes it feasible to prevent unwanted actuation of the system with the low possibility of the presence of the obstacle and to more securely actuate the system with the high possibility of the presence of the obstacle. This achieves appropriate actuation of the system.

It also becomes feasible to prevent incorrect detection of the obstacle with the low possibility of the presence of the obstacle and to securely detect the obstacle with the high possibility of the presence of the obstacle. This achieves appropriate recognition of the obstacle.

For example, where there is a metal bump or the like on the road ahead the vehicle, the radar 3 might detect it as an obstacle. Namely, if the detection area by the radar 3 is set ahead the vehicle on the assumption that there is no other vehicle or the like ahead the vehicle within the predetermined range on the basis of the image information, the radar 3 can recognize a metal bump or a metal plate on the road, as an obstacle with a high possibility.

In contrast to it, the obstacle recognition system and obstacle recognition method according to the present embodiment are arranged to set the determination threshold for the obstacle detection by the radar 3 at the high value when the possibility of the presence of the obstacle is determined to be low based on the image information. For this reason, they are able to prevent an object posing no impediment to drive of the vehicle, such as a metal bump, from being detected as an obstacle, and thus they are able to detect a real obstacle to drive of the vehicle, thus implementing appropriate detection of the obstacle.

Furthermore, the obstacle recognition system and obstacle recognition method according to the present embodiment are arranged to, with detection of an obstacle, expedite the activation of the collision avoidance or collision influence relief with the obstacle. This can raise the possibility of avoidance of the collision with the obstacle, or relieve the influence of the collision.

The present embodiment described the case where the criterion value was changed so as to make the obstacle less likely to be detected with increase in the estimated possibility of the absence of the obstacle on the basis of the image information and where the criterion value was changed so as to make the obstacle more likely to be detected with increase in the estimated possibility of the presence of the obstacle on the basis of the image information, but it is also possible to adopt a configuration wherein the collision avoidance or collision influence relief is activated earlier when the possibility of the presence of the obstacle is estimated to be high based on the image information than when the possibility of the presence of the obstacle is estimated not to be high. For example, in the flowchart of Fig. 2, the collision avoidance or collision influence relief is activated earlier when it is determined at S20 that the obstacle presence flag is set than when it is determined at S 18 that the obstacle absence flag is set. This can enhance the possibility of avoidance of a collision with the obstacle, or relieve the influence of the collision.


Anspruch[de]
Hinderniserkennungssystem (1) mit: einem Radar (3) zum Erfassen der Existenz eines Objekts um ein Fahrzeug (5); einer Bildaufnahmevorrichtung (2) zum Aufnehmen eines Bildes (30) des Objekts; und einem Hinderniserfassungselement (4) zum Erfassen eines Hindernisses (33) für die Bewegung des Fahrzeugs (5) auf der Basis einer Ausgabe (61) des Radars (3);

gekennzeichnet durch
ein Schwellwertänderungselement (4) zum Ändern eines Schwellwerts (62, 63) bei der Hinderniserfassung des Hinderniserfassungselements auf der Basis von einer Bildinformation des Bildes (30), welches durch die Bildaufnahmevorrichtung (2) aufgenommen worden ist, wobei bei der Hinderniserfassung des Hinderniserfassungselements (4) der Schwellwert (62, 62) mit der Ausgabe (61) des Radars (3) verglichen wird, um die Anwesenheit/Abwesenheit des Hindernisses (33) zu bestimmen.
Hinderniserkennungssystem (1) gemäß Anspruch 1, welches ferner ein Hindernisabschätzelement (4) zum Abschätzen der Anwesenheit/Abwesenheit des Hindernisses (33) auf der Basis der Bildinformation aufweist,

wobei das Schwellwertänderungselement (4) den Schwellwert (62, 63) auf der Basis einer Möglichkeit des Vorhandenseins des Hindernisses (33), welche durch das Hindernisabschätzelement (4) abgeschätzt wird, ändert.
Hinderniserkennungssystem (1) gemäß Anspruch 2, wobei das Hindernisabschätzelement (4) eine Erfassungsfläche (44, 48) in dem Bild (30) auf der Basis der Ausgabe (61) des Radars (3) festlegt und die Anwesenheit des Hindernisses (33) in der Erfassungsfläche (44, 48) auf der Basis der Bildinformation abschätzt. Hinderniserkennungssystem (1) gemäß Anspruch 3, wobei die Erfassungsfläche (44, 48), welche durch das Hindernisabschätzelement (4) festgelegt wird, auf der Basis einer Entfernung von dem Fahrzeug (5) zu dem Objekt und einer Relativposition des Objekts zu dem Fahrzeug (5) festgelegt wird. Hinderniserkennungssystem (1) gemäß einem der Ansprüche 2 bis 4 wobei das Hindernisabschätzelement (4) die Anwesenheit des Hindernisses (33) auf der Basis einer Kantendichtenverteilung, welche durch Bildverarbeitung des Bildes (30) erhalten wird, abschätzt. Hinderniserkennungssystem (1) gemäß einem der Ansprüche 2 bis 5, wobei das Hindernisabschätzelement (4) eine Möglichkeit der Abwesenheit des Hindernisses (33) auf der Basis der gesamten Kantendaten berechnet, welche durch Bildverarbeitung eines vorbestimmten Bildverarbeitungsbereichs (31) in dem Bild (30) erhalten wird, die Möglichkeit der Anwesenheit des Hindernisses (33) auf der Basis der Kantendaten berechnet, welche durch die Subtraktion der Kantendaten von Fahrbahnmarkierungen (32) von den gesamten Kantendaten resultiert, und die Anwesenheit des Hindernisses (33) auf der Basis der Möglichkeit der Abwesenheit des Hindernisses (33) und der Möglichkeit der Anwesenheit des Hindernisses (33) abschätzt. Hinderniserkennungssystem (1) gemäß einem der Ansprüche 2 bis 6, wobei das Schwellwertänderungselement (4) den Schwellwert (62) mit einer Zunahme der abgeschätzten Möglichkeit der Abwesenheit des Hindernisses (33) auf der Basis der Bildinformation erhöht, so dass das Hindernis (33) weniger wahrscheinlich erfasst wird. Hinderniserkennungssystem (1) gemäß einem der Ansprüche 2 bis 7, wobei das Schwellwertänderungselement (4) den Schwellwert (63) mit einer Erhöhung der abgeschätzten Möglichkeit der Anwesenheit des Hindernisses (33) auf der Basis der Bildinformation verringert, so dass das Hindernis (33) wahrscheinlicher erfasst wird. Hinderniserkennungssystem (1) gemäß einem der Ansprüche 2 bis 8, welches ferner ein Kollisionsvermeidungs-/Abhilfeelement (4) zum Vermeiden einer Kollision mit dem Hindernis (33) oder zum Abhelfen einer Kollisionsauswirkung aufweist,

wobei die Kollisionsvermeidung oder die Kollisionsauswirkungsabhilfe durch das Kollisionsvermeidungs-/Abhilfeelement (4) früher aktiviert wird, wenn das Hindernisabschätzelement (4) abschätzt, dass die Möglichkeit der Anwesenheit des Hindernisses (33) hoch ist, als wenn das Hindernisabschätzelement (4) abschätzt, dass die Möglichkeit der Anwesenheit des Hindernisses (33) nicht hoch ist.
Hinderniserkennungsverfahren mit: einem Hinderniserfassungsschritt (S26) zum Erfassen eines Hindernisses (33) für die Bewegung des Fahrzeugs auf der Basis einer Ausgabe (61) eines Radars (3) zum Erfassen der Existenz eines Objekts um ein Fahrzeug (5) unter Verwendung eines Schwellwerts (62, 63), einem Bildaufnahmeschritt zum Aufnehmen eines Bildes des Objekts;

gekennzeichnet durch
einen Schwellwertänderungsschritt (S22, S24) zum Ändern des Schwellwerts (62, 63) bei der Hinderniserfassung auf der Basis der Bildinformation des Bildaufnahmeschritts; wobei der Hinderniserfassungsschritt (S26) das Vergleichen des Schwellwerts (62, 63), welcher in dem Schwellwertänderungsschritt (S22, S24) festgelegt wird, mit der Ausgabe (61) des Radars (3) beinhaltet, um die Anwesenheit/Abwesenheit des Hindernisses (33) zu bestimmen.
Hinderniserkennungsverfahren gemäß Anspruch 10, welches ferner einen Hindernisabschätzschritt (S16) zum Abschätzen der Anwesenheit/Abwesenheit des Hindernisses (32) auf der Basis der Bildinformation aufweist,

wobei der Schwellwertänderungsschritt (S22, S24) das Ändern des Schwellwerts (62, 63) auf der Basis einer Möglichkeit der Anwesenheit des Hindernisses (33), welche in dem Hindernisabschätzschritt (S16) abgeschätzt wird, beinhaltet.
Hinderniserkennungsverfahren gemäß Anspruch 11, wobei der Hindernisabschätzschritt (S16) das Festlegen einer Erfassungsfläche (44, 48) in dem Bild (30) auf der Basis der Ausgabe des Radars (3) und das Abschätzen der Anwesenheit des Hindernisses (33) in der Erfassungsfläche (44, 48) auf der Basis der Bildinformation beinhaltet. Hinderniserkennungsverfahren gemäß Anspruch 12, wobei die Erfassungsfläche (44, 48) in dem Hindernisabschätzschritt (S16) auf der Basis einer Entfernung von dem Fahrzeug (5) zu dem Objekt und einer Relativposition des Objekts zu dem Fahrzeug (5) festgelegt wird. Hinderniserkennungsverfahren gemäß einem der Ansprüche 11 bis 13, wobei der Hindernisabschätzschritt (S16) das Abschätzen der Anwesenheit des Hindernisses (33) auf der Basis einer Kantendichtenverteilung, welche durch Bildverarbeitung des Bildes (30) erhalten wird, beinhaltet. Hinderniserkennungsverfahren gemäß einem der Ansprüche 11 bis 14, wobei der Hindernisabschätzschritt (S16) das Berechnen einer Möglichkeit der Abwesenheit des Hindernisses (33) auf der Basis der gesamten Kantendaten, welche durch Bildverarbeitung eines vorbestimmten Bereichs (31) in dem Bild (30) erhalten wird (S18), das Berechnen der Möglichkeit der Anwesenheit des Hindernisses (33) auf der Basis der Kantendaten, welche aus der Subtraktion der Kantendaten von Fahrbahnmarkierungen (32) von den gesamten Kantendaten resultieren (S20), und das Abschätzen der Anwesenheit des Hindernisses (33) auf der Basis der Möglichkeit der Abwesenheit des Hindernisses (33) und der Möglichkeit der Anwesenheit des Hindernisses (33) beinhaltet. Hinderniserkennungsverfahren gemäß einem der Ansprüche 11 bis 15, wobei der Schwellwertänderungsschritt (S24) das Erhöhen des Schwellwerts (62) mit einer Erhöhung in der abgeschätzten Möglichkeit der Abwesenheit des Hindernisses (33) auf der Basis der Bildinformation aufweist (S18), so dass das Hindernis (33) weniger wahrscheinlich erfasst wird (S18). Hinderniserkennungsverfahren gemäß einem der Ansprüche 11 bis 16, wobei der Schwellwertänderungsschritt (S22) die Verringerung des Schwellwerts (63) mit einer Zunahme der abgeschätzten Möglichkeit der Anwesenheit des Hindernisses (33) auf der Basis der Bildinformation aufweist (S20), so dass das Hindernis (33) wahrscheinlicher erfasst wird. Hinderniserkennungsverfahren gemäß einem der Ansprüche 11 bis 17, welches ferner einen Kollisionsvermeidungs-/Abhilfeschritt (S28) zum Vermeiden einer Kollision mit dem Hindernis (33) oder zum Abhelfen einer Kollisionsauswirkung aufweist,

wobei die Kollisionsvermeidung oder die Kollisionsauswirkungsabhilfe in dem Kollisionsvermeidungs-/Abhilfeschritt (S28) früher aktiviert wird, wenn die Möglichkeit der Anwesenheit des Hindernisses (33) in den Hindernisabschätzschritt (S16) als hoch eingeschätzt wird, als wenn die Möglichkeit der Anwesenheit des Hindernisses (33) als nicht hoch eingeschätzt wird.
Anspruch[en]
An obstacle recognition system (1) comprising: a radar (3) for detecting existence of an object around a vehicle (5); an image taking device (2) for taking an image (30) of the object; and an obstacle detecting element (4) for detecting an obstacle (33) to travel of the vehicle (5) on the basis of an output (61) from the radar (3); characterised by further comprising a threshold changing element (4) for changing a threshold (62, 63) in the obstacle detection of the obstacle detecting element on the basis of image information of the image (30) taken by the image taking device (2), wherein in the obstacle detection of the obstacle detecting element (4) the threshold (62, 63) is compared with the output (61) from the radar (3) to determine the presence/absence of the obstacle (33). The obstacle recognition system (1) according to Claim 1, further comprising an obstacle estimating element (4) for estimating presence/absence of the obstacle (33) on the basis of the image information,

wherein the threshold changing element (4) changes the threshold (62, 63) on the basis of a possibility of the presence of the obstacle (33) estimated by the obstacle estimating element (4).
The obstacle recognition system according to Claim 2, wherein the obstacle estimating element (4) sets a detection area (44, 48) in the image (30) on the basis of the output (61) from the radar (3) and estimates the presence of the obstacle (33) in the detection area (44, 48) on the basis of the image information. The obstacle recognition system according to Claim 3, wherein the detection area (44, 48) set by the obstacle estimating element (4) is set based on a distance from the vehicle (5) to the object and a relative position of the object to the vehicle (5). The obstacle recognition system (1) according to any one of Claims 2 to 4, wherein the obstacle estimating element (4) estimates the presence of the obstacle (33) on the basis of an edge density distribution obtained through image processing of the image (30). The obstacle recognition system (1) according to any one of Claims 2 to 5, wherein the obstacle estimating element (4) calculates a possibility of absence of the obstacle (33) on the basis of entire edge data obtained through image processing of a predetermined image processing region (31) in the image (30), calculates the possibility of the presence of the obstacle (33) on the basis of edge data resulting from subtraction of edge data of lane markers (32) from the entire edge data, and estimates the presence of the obstacle (33) on the basis of the possibility of the absence of the obstacle (33) and the possibility of the presence of the obstacle (33). The obstacle recognition system according to any one of Claims 2 to 6, wherein the threshold changing element (4) increases the threshold (62) with an increase in the estimated possibility of absence of the obstacle (33) based on the image information so that the obstacle (33) becomes less likely to be detected. The obstacle recognition system according to any one of Claims 2 to 7, wherein the threshold changing element (4) decreases the threshold (63) with an increase in the estimated possibility of presence of the obstacle (33) based on the image information so that the obstacle (33) becomes more likely to be detected. The obstacle recognition system according to any one of Claims 2 to 8, further comprising a collision avoiding/relieving element (4) for avoiding a collision with the obstacle (33) or for relieving influence of the collision,

wherein the collision avoidance or collision influence relief by the collision avoiding/relieving element (4) is activated earlier when the obstacle estimating element (4) estimates that the possibility of the presence of the obstacle (33) is high than when the obstacle estimating element (4) estimates that the possibility of the presence of the obstacle (33) is not high.
An obstacle recognition method comprising: an obstacle detecting step (S26) of detecting an obstacle (33) to travel of the vehicle on the basis of an output (61) from a radar (3) for detecting existence of an object around a vehicle (5) by use of a threshold (62, 63), an image taking step for taking an image of the object; characterised by further comprising a threshold changing step (S22, S24) of changing the threshold (62, 63) in obstacle detection on the basis of image information from the image taking step; wherein the obstacle detecting step (S26) comprises comparing the threshold (62, 63) set in the threshold changing step (S22, S24) with the output (61) from the radar (3) to determine the presence/absence of the obstacle (33). The obstacle recognition method according to Claim 10, further comprising an obstacle estimating step (S16) of estimating presence/absence of the obstacle (33) on the basis of the image information,

wherein the threshold changing step (S22, S24) comprises changing the threshold (62, 63) on the basis of a possibility of the presence of the obstacle (33) estimated in the obstacle estimating step (S16).
The obstacle recognition method according to Claim 11, wherein the obstacle estimating step (S16) comprises setting a detection area (44, 48) in the image (30) on the basis of the output from the radar (3) and estimating the presence of the obstacle (33) in the detection area (44, 48) on the basis of the image information. The obstacle recognition method according Claim 12, wherein the detection area (44, 48) in the obstacle estimating step (S 16) is set based on a distance from the vehicle (5) to the object and a relative position of the object to the vehicle (5). The obstacle recognition method according to any one of Claims 11 to 13, wherein the obstacle estimating step (S16) comprises estimating the presence of the obstacle (33) on the basis of an edge density distribution obtained through image processing of the image (30). The obstacle recognition method according to any one of Claims 11 to 14, wherein the obstacle estimating step (S16) comprises calculating a possibility of absence of the obstacle (33) on the basis of entire edge data obtained through image processing of a predetermined region (31) in the image (30) (S18), calculating the possibility of the presence of the obstacle (33) on the basis of edge data resulting from subtraction of edge data of lane markers (32) from the entire edge data (S20), and estimating the presence of the obstacle (33) on the basis of the possibility of the absence of the obstacle (33) and the possibility of the presence of the obstacle (33). The obstacle recognition method according to any one of Claims 11 to 15, wherein the threshold changing step (S24) comprises increasing the threshold (62) with an increase in the estimated possibility of absence of the obstacle (33) based on the image information (S18) such that the obstacle (33) becomes less likely to be detected (S 18). The obstacle recognition method according to either one of Claims 11 to 16, wherein the threshold changing step (S22) comprises decreasing the threshold (63) with an increase in the estimated possibility of presence of the obstacle (33) based on the image information (S20) such so that the obstacle (33) becomes more likely to be detected. The obstacle recognition method according to any one of Claims 11 to 17, further comprising a collision avoiding/relieving step (S28) of avoiding a collision with the obstacle (33) or relieving influence of the collision,

wherein the collision avoidance or collision influence relief in the collision avoiding/relieving step (S28) is activated earlier when the possibility of the presence of the obstacle (33) is estimated to be high in the obstacle estimating step (S16) than when the possibility of the presence of the obstacle (33) is estimated not to be high.
Anspruch[fr]
Système de reconnaissance d'obstacles (1) comprenant: un radar (3) pour détecter l'existence d'un objet autour d'un véhicule (5); un dispositif de prise d'images (2) pour prendre une image (30) de l'objet; et un élément de détection d'obstacles (4) pour détecter un obstacle (33) au déplacement du véhicule (5) sur la base d'une sortie (61) à partir du radar (3); caractérisé par le fait de comprendre en plus un élément de changement de seuil (4) pour changer un seuil (62, 63) dans la détection d'obstacles de l'élément de détection d'obstacles sur la base d'une information d'image de l'image (30) prise par le dispositif de prise d'images (2), où dans la détection d'obstacles de l'élément de détection d'obstacles (4) le seuil (62, 63) est comparé à la sortie (61) à partir du radar (3) pour déterminer la présence/l'absence de l'obstacle (33). Système de reconnaissance d'obstacles (1) selon la revendication 1, comprenant en plus un élément d'estimation d'obstacles (4) pour estimer une présence/absence de l'obstacle (33) sur la base de l'information d'image,

où l'élément de changement de seuil (4) change le seuil (62, 63) sur la base d'une possibilité de la présence de l'obstacle (33) estimée par l'élément d'estimation d'obstacles (4).
Système de reconnaissance d'obstacles selon la revendication 2, où l'élément d'estimation d'obstacles (4) établit une zone de détection (44, 48) dans l'image (30) sur la base de la sortie (61) à partir du radar (3) et estime la présence de l'obstacle (33) dans la zone de détection (44, 48) sur la base de l'information d'image. Système de reconnaissance d'obstacles selon la revendication 3, où la zone de détection (44, 48) établie par l'élément d'estimation d'obstacles (4) est établie sur la base d'une distance du véhicule (5) à l'objet et d'une position relative de l'objet par rapport au véhicule (5). Système de reconnaissance d'obstacles (1) selon l'une quelconque des revendications 2 à 4, où l'élément d'estimation d'obstacles (4) estime la présence de l'obstacle (33) sur la base d'une distribution de densité de bord obtenue à travers un traitement d'image de l'image (30). Système de reconnaissance d'obstacles (1) selon l'une quelconque des revendications 2 à 5, où l'élément d'estimation d'obstacles (4) calcule une possibilité d'absence de l'obstacle (33) sur la base d'une donnée de bord entière obtenue à travers un traitement d'image d'une région de traitement d'image prédéterminée (31) dans l'image (30), calcule la possibilité de la présence de l'obstacle (33) sur la base d'une donnée de bord résultant d'une soustraction d'une donnée de bord de marqueurs de voie (32) de la donnée de bord entière, et estime la présence de l'obstacle (33) sur la base de la possibilité de l'absence de l'obstacle (33) et la possibilité de la présence de l'obstacle (33). Système de reconnaissance d'obstacles selon l'une quelconque des revendications 2 à 6, où l'élément de changement de seuil (4) augmente le seuil (62) avec une augmentation de la possibilité estimée d'absence de l'obstacle (33) sur la base de l'information d'image de sorte qu'il devienne moins probable que l'obstacle (33) soit détecté. Système de reconnaissance d'obstacles selon l'une quelconque des revendications 2 à 7, où l'élément de changement de seuil (4) diminue le seuil (63) avec une augmentation de la possibilité estimée de présence de l'obstacle (33) sur la base de l'information d'image de sorte qu'il devienne plus probable que l'obstacle (33) soit détecté. Système de reconnaissance d'obstacles selon l'une quelconque des Revendications 2 à 8, comprenant en plus un élément évitant/dissipant une collision (4) pour éviter une collision avec l'obstacle (33) ou pour dissiper l'influence de la collision,

où le fait d'éviter une collision ou de dissiper l'influence d'une collision par l'élément évitant/dissipant une collision (4) est activé de manière plus précoce lorsque l'élément d'estimation d'obstacles (4) estime que la possibilité de la présence de l'obstacle (33) est élevée que lorsque l'élément d'estimation d'obstacles (4) estime que la possibilité de la présence de l'obstacle (33) n'est pas élevée.
Procédé de reconnaissance d'obstacles comprenant: une étape de détection d'obstacle (S26) pour détecter un obstacle (33) au déplacement du véhicule sur la base d'une sortie (61) à partir d'un radar (3) pour détecter l'existence d'un objet autour d'un véhicule (5) par l'utilisation d'un seuil (62, 63), une étape de prise d'image pour prendre une image de l'objet, caractérisé par le fait de comprendre en plus une étape de changement de seuil (S22, S24) pour changer le seuil (62, 63) dans une détection d'obstacles sur la base d'une information d'image de l'étape de prise d'images; où l'étape de détection d'obstacles (S26) comprend le fait de comparer le seuil (62, 63) établi à l'étape de changement de seuil (S22, S24) à la sortie (61) à partir du radar (3) pour déterminer la présence/l'absence de l'obstacle (33). Procédé de reconnaissance d'obstacles selon la revendication 10, comprenant en plus une étape d'estimation d'obstacle (S 16) pour estimer une présence/absence de l'obstacle (33) sur la base de l'information d'image,

où l'étape de changement de seuil (S22, S24) comprend le fait de changer le seuil (62, 63) sur la base d'une possibilité de la présence de l'obstacle (33) estimée à l'étape d'estimation d'obstacle (S 16).
Procédé de reconnaissance d'obstacles selon la revendication 11, où l'étape d'estimation d'obstacle (S 16) comprend le fait d'établir une zone de détection (44, 48) dans l'image (30) sur la base de la sortie à partir du radar (3) et estimer la présence de l'obstacle (33) dans la zone de détection (44, 48) sur la base de l'information d'image. Procédé de reconnaissance d'obstacles selon la revendication 12, où la zone de détection (44, 48) à l'étape d'estimation d'obstacle (S 16) est établie sur la base d'une distance du véhicule (5) à l'objet et d'une position relative de l'objet par rapport au véhicule (5). Procédé de reconnaissance d'obstacles selon l'une quelconque des revendications 11 à 13, où l'étape d'estimation d'obstacle (S 16) comprend le fait d'estimer la présence de l'obstacle (33) sur la base d'une distribution de densité de bord obtenue à travers un traitement d'image de l'image (30). Procédé de reconnaissance d'obstacles selon l'une quelconque des revendications 11 à 14, où l'étape d'estimation d'obstacle (S 16) comprend le fait de calculer une possibilité d'absence de l'obstacle (33) sur la base d'une donnée de bord entière à travers un traitement d'image d'une région prédéterminée (31) dans l'image (30) (S18), calculer la possibilité de la présence de l'obstacle (33) sur la base d'une donnée de bord résultant d'une soustraction d'une donnée de bord de marqueurs de voie (32) de la donnée de bord entière (S20), et estimer la présence de l'obstacle (33) sur la base de la possibilité de l'absence de l'obstacle (33) et la possibilité de la présence de l'obstacle (33). Procédé de reconnaissance d'obstacles selon l'une quelconque des revendications 11 à 15, où l'étape de changement de seuil (S24) comprend le fait d'augmenter le seuil (62) avec une augmentation de la possibilité estimée d'absence de l'obstacle (33) sur la base de l'information d'image (S 18) de manière à ce qu'il devienne moins probable que l'obstacle (33) soit détecté (S 18). Procédé de reconnaissance d'obstacles selon l'une quelconque des revendications 11 à 16, où l'étape de changement de seuil (S22) comprend le fait de diminuer le seuil (63) avec une augmentation de la possibilité estimée de présence de l'obstacle (33) sur la base de l'information d'image (S20) de manière à ce qu'il devienne plus probable que l'obstacle (33) soit détecté. Procédé de reconnaissance d'obstacles selon l'une quelconque des revendications 11 à 17, comprenant en plus une étape visant à éviter/dissiper une collision (S28) consistant à éviter une collision avec l'obstacle (33) ou à dissiper l'influence de la collision,

où le fait d'éviter une collision ou de dissiper l'influence d'une collision à l'étape visant à éviter/dissiper une collision (S28) est activé de manière plus précoce lorsque la possibilité de la présence de l'obstacle (33) est estimée comme étant élevée à l'étape d'estimation d'obstacles (S16) que lorsque la possibilité de la présence de l'obstacle (33) est estimée comme n'étant pas élevée.






IPC
A Täglicher Lebensbedarf
B Arbeitsverfahren; Transportieren
C Chemie; Hüttenwesen
D Textilien; Papier
E Bauwesen; Erdbohren; Bergbau
F Maschinenbau; Beleuchtung; Heizung; Waffen; Sprengen
G Physik
H Elektrotechnik

Anmelder
Datum

Patentrecherche

Patent Zeichnungen (PDF)

Copyright © 2008 Patent-De Alle Rechte vorbehalten. eMail: info@patent-de.com