Collision detection device and collision mitigation device

Abstract

Collision detection device (10) mounted on a separate vehicle and determining the probability of a collision with a moving object, wherein the collision detection device comprises: a collision determination device (S390) that determines whether a vehicle will collide with a moving object captured in a recorded image; a shielding determination device (S210 to S290, S330, S340, S370) that determines whether the moving object is in a shielded state or not, in which at least part of the moving object is hidden behind another object or the moving object appears from behind the other object; and a setting change device (S310, S350) which, when the moving object is in the shielded state, sets the time required by the collision determination device to complete a determination regarding the collision to a shorter time than when the moving object is not in the shielded state, a shielding object removal device (S230, S250) that removes at least one shielding object positioned in at least one specific area that is set as at least a sub-area in the captured image and that is suitable for shielding the moving object; and a motion object removal area setting means (S240, S260) that sets at least one motion object removal area to an area in which a field of view is determined to be shielded by the at least one shielding object that is positioned in a depth direction in the recorded image further than the at least one specific area from which the at least one shielding object was removed, wherein the shielding determination means determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object sampling area.

Description

BACKGROUND[Technical Area]

[0001] The present invention relates to a collision determining device and a collision mitigating device mounted on a self-vehicle, in which the collision determining device determines the probability of a collision with a moving object. [State of the art]

[0002] As the collision determination device described above, a configuration is known in which a warning is issued when a pedestrian moving behind a vehicle is detected (see, for example, JP 4 313 712 B).

[0003] In this collision determination device, the probability of a collision between a target object, such as a pedestrian, and the host vehicle must be determined early. However, if the probability of a collision is not accurately determined, erroneous operations such as false alarms occur more frequently and cause confusion. Therefore, false alarms are suppressed over time to perform collision determination in which the movement trajectory of the target object is accurately calculated.

[0004] In the presently described collision determination device according to JP 4 313 712 B, it is expected that the collision determination is preferentially performed in cases where the pedestrian moving behind a vehicle is visible. However, a certain amount of time is required to perform the collision determination described above. Therefore, the determination may not be made in a timely manner in cases where the target object suddenly appears from behind the shielding object, such as a vehicle.

[0005] DE 10 2004 018 813 A1 discloses a method for detecting and / or tracking objects that correspond to items in at least one detection zone of at least one sensor for electromagnetic radiation. Based on images of the detection zone repeatedly captured by the sensor, successive images are evaluated in successive cycles, and at least one object detected in one cycle based on a corresponding image is searched for in a subsequent cycle in a corresponding subsequent image in order to track the object.On the basis of the results of an at least preliminary evaluation of a current image during at least one current cycle, at least one part of a current image determined in the current cycle or an object recognized in the current cycle is assigned at least one part of a previous image and / or at least one indication relating to a previous state of the object or an article corresponding thereto, which is determined using a corresponding previous image in the current cycle.

[0006] DE 10 2009 035 072 A1 discloses a method for predicting or anticipating the position or location and / or movement of an object or vulnerable road user, for example, a pedestrian, relative to a vehicle. These methods are used to predict impending collisions between a vehicle and other vulnerable road users, such as pedestrians, cyclists, or even animals such as dogs. These methods are improved according to the invention by detecting the current movement data of the object from object-side sensors and transmitting them from the object-side transceiver unit to the vehicle-side transceiver unit, where they are used to predict the position and / or movement of the object. Furthermore, this prior art relates to device subunits for implementing the method according to the invention.

[0007] Further state-of-the-art information can be found in the publication by BROGGI, A. [et al.]: A new approach to urban pedestrian detection for automatic braking. In: IEEE Transactions on Intelligent Transportation Systems, Vol. 10, No. 4, 2009, pp. 594-605. - ISSN 1524-9050. SUMMARY OF THE INVENTION

[0008] Therefore, it is an object of the present invention to provide a collision determination device, a collision mitigation device, and a collision determination method, wherein the collision determination device and the collision mitigation device are mounted on a self-contained vehicle, in which the collision determination device detects the probability of collision with a moving object, and wherein these devices and the method are suitable for detecting a target object appearing from behind a shielding object early while reducing false alarms as much as possible.

[0009] This object is achieved by the collision determination device having the features of claim 1, by the collision mitigation device having the features of claim 11, and by the collision determination method of claim 12. Further advantageous developments and embodiments of the invention are the subject of the subsequent claims.

[0010] An exemplary embodiment provides a collision determination device mounted on a host vehicle that determines a probability of a collision of the host vehicle with a moving object. The collision determination device includes collision determination means, shielding determination means, and attitude changing means. The collision determination means determines whether or not the host vehicle will collide with a moving object captured in a captured image. The shielding determination means determines whether or not the moving object is in a shielded state in which at least a part of the moving object is hidden behind another object or the moving object appears from behind another object.When the moving object is in the shielded state, the setting changing means sets a time period required for the collision determining means to complete a determination regarding the collision shorter than in a case where the moving object is not in the shielded state.

[0011] According to the present collision determination device, when the moving object is in the shielded state, the time required to complete the determination regarding a collision with the moving object can be shortened. This allows for earlier determination of whether a collision will occur. When the moving object is not in the shielded state, it takes longer to determine a collision compared to a case where the moving object is in the shielded state. This can prevent erroneous determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The accompanying drawings show: Fig. 1 is a block diagram showing an overall configuration of an accident prevention safety system to which a collision mitigation device according to an embodiment is applied; Fig. 2 is a flowchart of a collision mitigation process performed by a main processing unit (CPU) of a collision mitigation controller; Fig. 3 a flowchart of an intersection determination process in the collision mitigation process shown in Fig. 2 is shown; Fig. 4 is a bird's eye view of vehicle detection areas and pedestrian detection areas according to the embodiment; Fig. 5 a bird's eye view of an example of a pedestrian's movement behavior; Fig. 6 is a flowchart of an actuation determination process in the collision mitigation process shown in Fig. 2; and Fig. 7 a bird's eye view of the vehicle detection areas and the pedestrian detection areas according to another example. DESCRIPTION OF THE EMBODIMENTS

[0013] Hereinafter, a collision determination apparatus and a collision shielding apparatus according to an embodiment will be described with reference to the drawings.

[0014] As in Fig. As shown in Figure 1, the collision mitigation device according to the present embodiment is applied to a pre-crash safety system (hereinafter referred to as PCS) 1. This PCS 1 is a system installed in a vehicle, such as a passenger car. The PCS 1 detects the risk of a vehicle collision and avoids a collision of the vehicle. In addition, the PCS 1 mitigates damage before a collision of the vehicle. More specifically, the PCS 1 includes, as shown in Fig. 1, a collision mitigation controller 10, various sensors 30, and a controlled object 40. The collision determination apparatus of the present embodiment is applied to the collision mitigation controller 10.

[0015] The various sensors 30 include, for example, a camera sensor 31, a radio sensor 32, a yaw rate sensor 33, and a wheel speed sensor 34. The camera sensor 31 is configured, for example, as a stereo camera suitable for detecting the distance to a target object. The camera sensor 31 detects the shape of the target object and the distance to the target object based on captured images. The target object is, for example, a pedestrian at a roadside obstacle or another vehicle captured in the images.

[0016] The wireless sensor 32 detects a target object and the position of the target object (relative position to the host vehicle). The yaw rate sensor 33 is configured as a known yaw rate sensor that detects the yaw rate of the vehicle.

[0017] The wheel speed sensor 34 detects the rotational frequency of the wheel, ie, the traveling speed of the vehicle. The detection results from the various sensors 30 are obtained by the collision mitigation controller 10.

[0018] The camera sensor 31 and the wireless sensor 32 detect target objects positioned in the direction of travel of the vehicle at a predetermined interval (such as 100 ms) set in advance. Furthermore, the wireless sensor 32 detects the shape and size of the target object by transmitting electromagnetic waves directed toward the target object and receiving reflected waves from the transmitted electromagnetic waves.

[0019] The collision mitigation controller 10 is configured as a conventional computer. The computer includes a main processing unit (CPU) 11, a read-only memory (ROM) 12, a random access memory (RAM) 13, and the like. The collision mitigation controller 10 executes a program stored in the ROM 12 based on the detection results from the various sensors 30 and the like. The collision mitigation controller 10 thereby performs various processes, such as the collision mitigation process described below.

[0020] The collision mitigation controller 10 performs such processing and operates the controlled object 40 based on the processing results of the processing. The controlled object 40 includes, for example, an actuator that operates a brake, a steering wheel, a seat belt, or the like, and a warning device that issues a warning. According to the present embodiment, an exemplary case in which the controlled object 40 is the brake will be described below.

[0021] As described above, when the CPU 11 operates the automatic brake function, the CPU 11 operates the controlled object 40 to achieve a deceleration rate and an acceleration amount (the speed difference before and after the automatic brake operation) set in advance based on a detection signal from the wheel speed sensor 34.

[0022] Next, a collision mitigation process is performed with respect to Fig. 2 and the following drawings. The collision mitigation process is performed when automatic braking is performed. The collision mitigation process is started at a predetermined interval (e.g., approximately 50 ms) that is set in advance.

[0023] More specifically, in the collision mitigation process, as in Fig. 2, the CPU 11 of the collision mitigation controller 10 first inputs information regarding a target object (step S100). In this process processing operation, the CPU 11 acquires the most recent information regarding the position of the target object detected by the camera sensor 31 and the wireless sensor 32.

[0024] Subsequently, the CPU 11 performs detection of the target object (step S110). In this processing, the type of the target object (such as a vehicle, a pedestrian, a bicycle, or a motorcycle) is detected (such as through pattern recognition) based on the shape and the like of the target object detected by the camera sensor 31. A target object previously recorded in the RAM 13 or the like and the target object detected at this time are then associated with each other.

[0025] Next, the CPU 11 performs an intersection determination process (step S120). In the intersection determination process, it is determined whether or not a moving object will cross in front of the host vehicle in the traveling direction.

[0026] As in Fig. As shown in Figure 3, in the intersection determination process, the CPU 11 first obtains the vehicle speed and the relative speed to the target object (step S200). The relative speed can be determined using the Doppler effect that occurs when the wireless sensor 32 detects the target object or from the position history of the target object (relative movement history).

[0027] Next, the CPU 11 sets two areas on the left and right sides in front of the own vehicle as vehicle detection areas (corresponding to at least one specific area) (steps S210 and S220). In this process, as shown in Fig. 4, the vehicle detection areas (corresponding to a left specific area and a right specific area) 51 and 53 are set into areas in which stopped vehicles 61 to 63 are assumed to be present, in the direction of travel of (in front of) a host vehicle 100. The vehicle detection areas 51 and 53 are divided into areas on the left side and the right side.

[0028] The positions and sizes of the vehicle detection areas 51 and 53 are set based on the traveling speed of the own vehicle or the relative speed of the stopped vehicles 61 to 63 (shielding objects). For example, in a case where the traveling speed or relative speed is 20 km / h, the position of each vehicle detection area 51 and 53 is set to a position (size is 10 m in depth) that is 5 m to 15 m away from the own vehicle 100. As the traveling speed or relative speed increases, the position of each vehicle detection area 51 and 53 becomes farther away from the own vehicle 100. Furthermore, the size (depth) of each vehicle detection area 51 and 53 becomes larger.

[0029] Next, the CPU 11 judges whether or not a stopped vehicle is detected in the vehicle detection area 51 on the left side (step S230). The stopped vehicle is a vehicle moving at a speed at which the vehicle can be considered stopped (e.g., a vehicle moving at a speed of +20 km / h to less than -20 km / h, or moving at a very low speed; the moving speed here refers to the absolute speed). If it is judged that no stopped vehicle is detected in the vehicle detection area 51 on the left side (NO at step S230), the CPU 11 proceeds to step S250.

[0030] If it is judged that a stopped vehicle is detected in the vehicle detection area 51 on the left side (YES at step S230), the CPU 11 generates a pedestrian detection area (corresponding to at least a moving object extraction area) 52 on the left side in the traveling direction of the own vehicle (step S240). Here, the pedestrian detection area 52 is set to an area in which a field of view is assumed to be blocked by the stopped vehicle. The pedestrian detection area 52 is set further in the depth direction of the captured image than the vehicle detection area 51 in which the stopped vehicle was detected.

[0031] The pedestrian detection area 52 is set such that the starting point is a position further moved in the depth direction by a distance corresponding to the length of the vehicle with respect to the position of the stopped vehicle (detection position). The position at the end point in the depth direction (size of the pedestrian detection area 52) is set depending on the traveling speed of the host vehicle or the relative speed of the pedestrian. Similar to the vehicle detection areas 51 and 53, the pedestrian detection area 52 is also set to become larger as the traveling speed of the host vehicle or the relative speed of the pedestrian increases.

[0032] Next, the CPU 11 judges whether a stopped vehicle is detected in the vehicle detection area 53 on the right side (step S250). If it is judged that no stopped vehicle is detected in the vehicle detection area 53 on the right side (NO in step S250), the CPU 11 proceeds to step S270.

[0033] If it is judged that a stopped vehicle is detected in the vehicle detection area 53 on the right side (YES at step S250), the CPU 11 creates a pedestrian detection area 54 on the right side (step S260). In this processing, processing similar to that for creating the pedestrian detection area 52 on the left side is performed.

[0034] Through the processing in steps S230 to S260, when a stopped vehicle is detected in the vehicle detection area 51 on the left side, the pedestrian detection area 52 is set on the left side in the traveling direction of the own vehicle. When a stopped vehicle is detected in the vehicle detection area 53 on the right side, the pedestrian detection area 54 is set on the right side in the traveling direction of the own vehicle.

[0035] Furthermore, a pedestrian 60 located within the pedestrian detection area 52 or 54 can be said to be in a shielded state. In the shielded state, at least a portion of the pedestrian 60 is hidden behind the stopped vehicle. Alternatively, the pedestrian 60 may appear from behind the stopped vehicle.

[0036] According to the present embodiment, when the stopped vehicles 62 and 63 (see Fig. 4) When pedestrian detection areas 51 and 53 are detected, the pedestrian detection areas 52 and 54 are set with respect to the position of the stopped vehicle 62, which is closest to the own vehicle, among the stopped vehicles 62 and 63. Once the pedestrian detection areas 52 and 54 are set, the pedestrian detection areas 52 and 54 remain set until the own vehicle directly passes the pedestrian detection areas 52 and 54 (until the own vehicle moves at a distance from the position where the pedestrian detection areas 52 and 54 are set to the end position of the end point in the depth direction [moving object extraction distance]).

[0037] Next, the CPU 11 judges whether a stopped vehicle is detected in at least one of the vehicle detection areas 51 and 53 on the left side and the right side (step S270). If it is judged that a stopped vehicle is detected (YES at step S270), the CPU 11 judges whether or not a pedestrian is detected in the pedestrian detection area 52 on the left side (step S280). If it is judged that no pedestrian is detected (NO at step S280), the CPU 11 proceeds to step S330 as described below.

[0038] When it is judged that a pedestrian is detected (YES at step S280), the CPU 11 judges whether or not a distance from the position at which the stopped vehicle is detected to the position at which the pedestrian is detected is a reference distance set in advance (a distance used to detect a pedestrian who is close to the stopped vehicle in the shielded state and has a higher risk) (step S290).

[0039] When it is judged that the distance from the position at which the stopped vehicle is detected to the position at which the pedestrian is detected is a reference distance (YES at step S290), the CPU 11 shortens the time required to perform a cross movement determination (determination of whether or not the pedestrian will cross in front of the own vehicle) of the pedestrian (step S310).

[0040] Specifically, the time required for the collision determination to be completed is relaxed or extended to a short period of time by setting a reference condition. The reference condition is used when determining a collision. The reference condition indicates, for example, the number of images (number of frames) used when determining the trajectory of a moving object, the movement distance (absolute value) in the transverse direction of a moving object, and the like.

[0041] In the case where the reference condition is the number of frames, relaxing the reference condition corresponds to reducing the number of frames. In the case where the reference condition is the movement distance, relaxing the reference condition corresponds to reducing the distance value. This allows the lateral movement determination to be completed sooner.

[0042] If the reference condition is changed during this process, the reference condition will be further relaxed as the transverse distance from the position of the own vehicle to the position of the detected moving object becomes smaller. For example, as shown in Fig. 4, with regard to the transverse distance of the host vehicle 100, the transverse distance from the host vehicle 100 to the stopped vehicles 62 and 63 on the right side is greater than the transverse distance from the host vehicle 100 to the stopped vehicle 61 on the left side.

[0043] In such a situation, the reference condition is more relaxed with respect to the pedestrian 60 appearing from behind the stopped vehicle 61. The transverse distance of this pedestrian 60 is shorter than that of a pedestrian appearing from behind the stopped vehicle 62.

[0044] This is used to determine the amount of transverse or lateral movement of the moving object, as in Fig. 5, the movement path of the pedestrian with respect to the own vehicle 100 is used. In the Fig. In the example shown in Figure 5, images with five frames from t=X to (X+4n) are used to more accurately determine the amount of movement of the moving object. However, if the reference condition is relaxed, images with three frames from t=X to (X+2n) are used, for example.

[0045] Next, at step S290, when it is judged that the distance from the position at which the stopped vehicle is detected to the position at which the pedestrian is detected is not a reference distance (NO at step S290), the CPU 11 considers the lateral movement determination of the pedestrian as an ordinary state in which the amount of time required to perform the lateral movement determination is not shortened (step S320).

[0046] Subsequently, the CPU 11 performs processing similar to the processing (steps S280 to S320) for the left-side pedestrian detection area 53 and the right-side pedestrian detection area 54 (steps S330 to S360). When such processing is completed, the CPU 11 proceeds to step S390 as described below.

[0047] If it is judged at step S270 that no stopped vehicle is detected (NO at step S270), the CPU 11 judges whether or not a pedestrian is detected in the predetermined area by each sensor (step S370). If it is judged that a pedestrian is detected (YES at step S370), the CPU 11 sets the pedestrian lateral movement determination to a normal state in which the time required to perform the lateral movement determination is not shortened (step S380). The CPU 11 then proceeds to step S390.

[0048] If it is judged that no pedestrian is detected (NO at step S370), the CPU 11 proceeds to step S390. At step S390, the CPU 11 performs intersection determination based on the setting (step S390). As the threshold (reference condition) and the like used to perform intersection determination, the setting in which the required time is shortened, the ordinary state setting in which the required time is not shortened, and the like are used.

[0049] Then, based on whether or not a parameter value (such as the relative speed, the relative distance, or the amount of lateral movement) regarding the positional relationship between the pedestrian and the own vehicle satisfies the reference condition set in advance, it is determined whether or not the pedestrian detected in the captured image will cross in front of the own vehicle.

[0050] When such process processing is completed, the CPU 11 executes the process processing flow in Fig. 2 and performs an operation determination process (step S130). In the operation determination process, whether or not it is time to operate the controlled object 40 is determined based on an assumed travel path of the target object, the distance to the target object, the relative speed to the target object, and the like. When it is time to operate the controlled object 40, an operation instruction is generated and recorded in the RAM 13.

[0051] In the operation determination process, the CPU 11 calculates as shown in Fig. 6, a collision time is calculated based on the behavior of the target object and the relative speed to the target object (step S410). The collision time indicates the time until the own object will collide with the target object.

[0052] Subsequently, the CPU 11 calculates a collision probability (step S420). The collision probability indicates the probability of a collision between the host vehicle and the target object. Here, a plurality of correction coefficients are calculated for the collision probability based on the intersection determination result described above, the collision time, the speed of the moving object, the host vehicle's speed or relative speed, the positional relationship, and the like.

[0053] The collision probability is then obtained by a calculation performed using the correction coefficients. Based on the intersection determination result, the collision probability is set to a higher value when it is determined that the pedestrian will cross in front of the vehicle than when it is determined that the pedestrian will not cross in front of the vehicle.

[0054] Subsequently, the CPU 11 compares the collision probability with a threshold value set in advance (step S440). If it is judged that the collision probability is equal to or higher than the threshold value (YES in step S440), the CPU 11 generates an automatic braking operation instruction (i.e., sets a flag in the RAM 13) (step S450). Subsequently, the CPU 11 terminates the operation determination process.

[0055] If it is judged that the collision probability is lower than the threshold (YES at step S440), the CPU 11 terminates the operation determination process. When the operation determination process is completed, the CPU 11 executes the process flow in Fig. 2 and performs a decision process (step S140).

[0056] In the decision process, it is finally determined whether the controlled object 40 is actually operated or not. Specifically, in a case where the operation instruction for automatic braking is recorded in the RAM 13 in the operation determination process, if the driver is performing a collision avoidance maneuver and there is sufficient margin until a collision with the target object, it is considered that the driver can perform collision avoidance himself.

[0057] This prevents the automatic braking from being applied. This means that the driver's intervention is prioritized in the decision-making process if the collision can be avoided. The automatic braking can be canceled or cancelled.

[0058] Next, the CPU 11 performs an operation control process (step S150). In the operation control process, the CPU 11 transmits the operation instruction corresponding to the controlled object 40 (to the respective controlled objects 40 if there are a plurality of controlled objects 40) based on the generated operation instruction (flag).

[0059] When such an actuation control process is completed, the collision mitigation process is also completed.

[0060] In the PCS 1 described in detail above, the collision mitigation controller 10 determines the probability of a collision between the host vehicle and the target object. If the probability of a collision is higher than a predetermined threshold, the collision mitigation controller 10 actuates an actuator to avoid a collision. Furthermore, the collision mitigation controller 10 determines whether or not the host vehicle will collide with the moving object (pedestrian) detected in a captured image.

[0061] Next, it is determined whether the moving object is in a shielded state or not. In the shielded state, at least a part of the moving object is hidden behind another object. Alternatively, the moving object appears from behind the other object. Further, the collision mitigation controller 10 sets the time required until the determination regarding the collision (the intersection determination process according to the present embodiment, but may also be another process) is completed when the moving object is in a shielded state to a shorter time than when the moving object is not in a shielded state.

[0062] According to such a PCS 1, when the moving object is in the shielded state, the time required to complete the determination regarding the collision with the moving object can be shortened. Therefore, it can be determined early whether a collision will occur. On the other hand, when the moving object is not in the shielded state, it takes longer to determine the collision compared to a case where the moving object is in the shielded state. This can prevent erroneous determination.

[0063] Furthermore, in the PCS 1 described above, the collision mitigation controller 10 judges whether the host vehicle will collide with a moving object detected in a captured image by determining whether a parameter value regarding the positional relationship between the moving object and the host vehicle satisfies a reference condition set in advance. The collision mitigation controller 10 relaxes the reference condition used to determine the collision, thereby setting the time required until the collision determination is completed to a short period of time.

[0064] According to this PCS 1, the reference condition is relaxed. Therefore, the parameter value regarding the positional relationship between the moving object and the host vehicle can more easily satisfy the reference condition at an early stage. This can shorten the time required to complete the collision determination.

[0065] Further, in the PCS 1 described above, the collision mitigation controller 10 extracts a shielding object that can shield the moving object and is positioned within the vehicle detection areas 51 and 53.

[0066] The vehicle detection areas 51 and 53 are set as different areas in the captured image. The collision mitigation controller 10 then sets the pedestrian detection areas 52 and 54 to areas where the field of view is determined to be obstructed by the obstructing object.

[0067] The pedestrian detection areas 52 and 54 are further set in the depth direction in the captured image as the vehicle detection areas 51 and 53 from which the shielding object has been extracted. Furthermore, when the moving object is detected in the pedestrian detection areas 52 and 54, the moving object is determined to be in the shielded state.

[0068] According to the present PCS 1, the moving object is determined to be in the shielded state when the moving object is detected in the pedestrian detection areas 52 and 54. Therefore, it can be easily determined whether the moving object is in the shielded state or not.

[0069] In addition, in the PCS 1 described above, the collision mitigation controller 10 determines that the moving object is in the shielded state when the moving object is detected in the pedestrian detection areas 52 and 54 during a period from when the shielding object is removed until when the own vehicle moves by the preset removed distance of the moving object.

[0070] According to the present PCS 1, the pedestrian detection areas 52 and 54 that were previously set can be maintained until the host vehicle moves the extracted distance of the moving object, even if the pedestrian detection areas 52 and 54 move over the elapsed time. Therefore, collision determination regarding the moving object detected in this area can be performed quickly.

[0071] Further, according to the above-described PCS 1, the collision mitigation controller 10 adjusts the positions and sizes of the vehicle detection areas 51 and 53 based on the traveling speed of the own vehicle or the relative speed of the shielding object.

[0072] According to the present PCS 1, the positions and sizes of the vehicle detection areas 51 and 53 can be adjusted to take into account that the size of the area to be focused changes depending on the traveling speed of the host vehicle or the relative speed of the shielding object. This can improve safety.

[0073] When this configuration is used, the vehicle detection areas 51 and 53 can be adjusted after the shielding object is removed. Then, it can be determined whether the shielding object is positioned within the vehicle detection areas 51 and 53 or not.

[0074] Furthermore, in the PCS 1 described above, the collision mitigation controller 10 adjusts the positions and sizes of the pedestrian detection areas 52 and 54 based on the traveling speed of the own vehicle or the relative speed of the moving object.

[0075] In such a PCS 1, the positions and sizes of the pedestrian detection areas 52 and 54 can be adjusted taking into account the size of the area to be processed in an early stage with respect to the moving object, which changes depending on the traveling speed of the host vehicle or the relative speed of the moving object. Thus, safety can be improved.

[0076] Further, in the PCS 1 described above, the collision mitigation controller 10 sets the pedestrian detection areas 52 and 54 with respect to the position of the shielding object that is closest to the own vehicle among the shielding objects within the vehicle detection areas 51 and 53.

[0077] According to the present PCS 1, the collision determination with respect to the moving object appearing from behind the shielding object can be quickly performed.

[0078] In addition, in the PCS 1 described above, the vehicle determination areas 51 and 53 are set on the left side and the right side in the direction of travel of the own vehicle.

[0079] According to the present PCS 1, the shielding objects and the moving objects can be detected for each vehicle detection area 51 and 53.

[0080] Furthermore, in the PCS 1 described above, when the shielding object is removed from the vehicle detection area 51 on the left side, the collision mitigation controller 10 sets the pedestrian detection area 52 on the left side in the traveling direction of the host vehicle. When the shielding object is removed from the vehicle detection area 53 on the right side, the collision mitigation controller 10 sets the pedestrian detection area 54 on the right side in the traveling direction of the host vehicle.

[0081] According to the present PCS 1, it can be identified whether the detection position of the moving object is on the left side or the right side.

[0082] Furthermore, in the PCS 1 described above, when the moving object is in the shielded state, the collision mitigation controller 10 sets the time required until the determination regarding the collision is completed to a shorter time as the distance in the lateral direction from the position of the own vehicle to the position of the detected moving object becomes smaller.

[0083] According to the present PCS 1, the collision can be determined earlier for a moving object that is closer to the vehicle with respect to the direction of travel of the own vehicle and for which the probability of collision is high. [Other embodiments]

[0084] The present disclosure is not limited by the embodiment described above.

[0085] Furthermore, according to the present disclosure, a section may be omitted in an embodiment as long as the present problem can be solved. Furthermore, an embodiment in which several embodiments as described above are combined is also conceivable. Furthermore, each embodiment belongs to the scope of the present disclosure as long as it is feasible and does not deviate from the scope of the present disclosure and the appended claims.

[0086] Furthermore, although reference numerals have been used in the description of the embodiment according to the scope of the claims, the reference numerals should be understood only for the purpose of simplifying the understanding of the present disclosure according to the claims and should not limit the technical scope of the present disclosure according to each claim.

[0087] For example, according to the above-described embodiment, the collision mitigation controller 10 determines the pedestrian to be in the shielded state when the pedestrian is detected in the pedestrian detection areas 52 and 54 and the position of the stopped vehicle and the position of the pedestrian are within a reference distance. However, the pedestrian may also be determined to be in the shielded state when it is detected that the pedestrian is in the pedestrian detection areas 52 and 54.

[0088] Furthermore, according to the above-described embodiment, the range over which image processing is performed with respect to the image captured by the camera sensor 31 and the range over which the wireless sensor 32 performs scanning are not specified. Therefore, the scanning range can be set to any range, such as the entire area. However, the range over which the target object is extracted can also be limited to the vehicle detection areas 51 and 53 and the pedestrian detection areas 52 and 54. This can reduce the computational load for extracting the target object.

[0089] Furthermore, according to the present embodiment, a configuration is proposed in which detection accuracy of the target object is improved by using both the camera sensor 31 and the wireless sensor 32.

[0090] However, the present embodiment can also be realized according to a configuration including either the camera sensor 31 or the wireless sensor 32.

[0091] Furthermore, according to the above-described PCS 1, the collision mitigation controller 10 sets the pedestrian detection areas 52 and 54 to be maintained from the time the shielding object is removed until the host vehicle passes through the pedestrian detection areas 52 and 54. The pedestrian detection areas 52 and 54 can be maintained until a moving object removal time, which is set in advance, has elapsed.

[0092] According to the present PCS 1, the pedestrian detection areas 52 and 54 that were previously set can be maintained until the moving object extraction time has passed, even if the pedestrian detection areas 52 and 54 move over the elapsed time. Therefore, collision determination for a moving object detected in these areas can be performed quickly.

[0093] In addition, the PCS 1 described above, as for example in Fig. 7, the pedestrian detection areas 52 and 54 may be set for a shielding object (a roadside object 65), such as a building or a tree, which shields the moving object, such as a pedestrian 60 or a bicycle, instead of a vehicle.

[0094] The PCS 1 corresponds to a collision mitigation device according to the exemplary embodiment. The collision mitigation controller 10 corresponds to a collision determination device of the exemplary embodiment. The processing at step S120 corresponds to a collision detection means according to the exemplary embodiment. The processing at steps S130 to S150 corresponds to the collision avoidance means of the exemplary embodiment.

[0095] Furthermore, the processing in steps S200 to S220 corresponds to a specific area setting means according to the exemplary embodiment. The processing in steps S240 and S260 corresponds to a moving object extraction area setting means or a pedestrian area setting means of the exemplary embodiment. The processing in steps S230 and S250 corresponds to a shielding object extraction means of the exemplary embodiment.

[0096] Furthermore, the processing at steps S310 and S350 corresponds to the setting changing means of the exemplary embodiment. The processing at steps S210 to S290, S330, S340, and S370 corresponds to the shielding determination means of the exemplary embodiment. The processing at step S390 corresponds to the collision determination means of the exemplary embodiment.

[0097] The collision determination device (collision mitigation controller 10) can be applied to a collision determination program in such a manner as to allow a computer to realize the means that configures the collision determination device.

[0098] Furthermore, the elements of the collision determination device (collision mitigation controller 10) and the elements of the collision mitigation device (PCS 1) can be selectively combined as needed. In this case, specific configurations may be omitted within the scope of the present disclosure.

Claims

[1] A collision determination device (10) mounted on a vehicle of the vehicle itself and determining a probability of collision with a moving object, the collision determination device comprising: a collision determination means (S390) that determines whether or not a host vehicle will collide with a moving object detected in a captured image; a shielding determination means (S210 to S290, S330, S340, S370) that determines whether or not the moving object is in a shielded state in which at least a part of the moving object is hidden behind another object or the moving object appears from behind the other object; and a setting changing means (S310, S350) which, when the moving object is in the shielded state, sets a time period required for the collision determination means to complete a determination regarding the collision to a shorter time period than when the moving object is not in the shielded state, a shielding object extraction means (S230, S250) that extracts at least one shielding object positioned in at least one specific area set as at least one partial area in the captured image and which is suitable for shielding the moving object; and a moving object extraction area setting means (S240, S260) that sets at least one moving object extraction area to an area in which a field of view is determined to be shielded by the at least one shielding object positioned further in a depth direction in the captured image than the at least one specific area from which the at least one shielding object has been extracted, wherein the shielding determining means determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object extraction area. [2] Collision determination device according to claim 1, wherein the collision determination means determines whether or not the own vehicle will collide with the moving object detected in the captured image based on whether or not a parameter value regarding a positional relationship between the moving object and the own vehicle satisfies a predetermined reference condition; and the setting changing means sets a time period required until the collision determining means completes a determination regarding the collision to a shorter time period by relaxing the predetermined reference condition when the collision determining means determines the collision. [3] A collision determination device according to claim 1 or 2, wherein the shielding determination means determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object extraction area during a period from when the at least one shielding object is extracted until when the own vehicle moves a predetermined extracted distance from the moving object. [4] The collision determination device according to claim 1 or 2, wherein the shielding determination means determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object extraction area during a period from when the at least one shielding object is extracted until when a predetermined moving object extraction time has elapsed. [5] Collision determination device according to one of claims 1 to 4, further comprising: a specific area setting means (S200 to S220) that sets a position and a size of the at least one specific area based on a traveling speed of the own vehicle or a relative speed of the at least one shielding object. [6] The collision determination device according to any one of claims 1 to 5, wherein the moving object extraction area setting means sets a position and a size of the at least one moving object extraction area based on a traveling speed of the own vehicle or a relative speed to the at least one shielding object. [7] The collision determination device according to any one of claims 1 to 6, wherein the moving object extraction area setting means sets the moving object extraction area with reference to a position of a shielding object that is closest to the own vehicle among the shielding objects in the specific area. [8] A collision determination device according to any one of claims 1 to 7, wherein the at least one specific region comprises: a left specific area, which is set on a left side in the direction of travel of the own vehicle; and a right-specific area that is set on a right side in the direction of travel of the vehicle. [9] The collision determination device according to claim 8, wherein the moving object extraction area setting means sets the at least one moving object extraction area on the left side in the traveling direction of the own vehicle when the at least one shielding object is extracted from the left specific area, and sets the at least one moving object extraction area on the right side in the traveling direction of the own vehicle when the at least one shielding object is extracted from the right specific area. [10] A collision determination device according to claim 8 or 9, wherein when the moving object is in the shielded state, the setting changing means sets a time period required until the collision determination means makes a determination regarding the collision to a shorter time period as a distance in the lateral direction from the position of the own vehicle to a position of the detected moving object becomes smaller. [11] A collision mitigation device (1) mounted on an own vehicle and mitigating collision damage when a probability of collision of the own vehicle with a moving object is high, the collision mitigation device comprising: a collision detection means (S120) that determines a probability of a collision of the own vehicle with a moving object; and a collision avoidance means (S130 to S150) that actuates an actuator capable of preventing the collision if the probability is higher than a predetermined threshold, wherein the collision determining means is configured by the collision determining device (10) mounted on a self-vehicle and determines a probability of a collision with a moving object, the collision determining device comprising: a collision determination means (S390) that determines whether or not a host vehicle will collide with a moving object detected in a captured image; a shielding determination means (S210 to S290, S330, S340, S370) that determines whether or not the moving object is in a shielded state in which at least a part of the moving object is hidden behind another object or the moving object appears from behind the other object; a setting changing means (S310, S350) which, when the moving object is in a shielded state, sets a time period required until the collision determining means completes a determination regarding the collision to a shorter time period than when the moving object is not in the shielded state, a shielding object extraction means (S230, S250) that extracts at least one shielding object positioned in at least one specific area set as at least one partial area in the captured image and which is suitable for shielding the moving object; and a moving object extraction area setting means (S240, S260) that sets at least one moving object extraction area to an area in which a field of view is determined to be shielded by the at least one shielding object positioned further in a depth direction in the captured image than the at least one specific area from which the at least one shielding object has been extracted, wherein the shielding determining means determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object extraction area. [12] A collision determination method comprising: for a collision determination device mounted on a vehicle that determines the probability of a collision with a moving object: Determining whether or not an own vehicle will collide with a moving object detected in a captured image; Determining whether or not the moving object is in a shielded state, wherein at least a part of the object to be moved is hidden behind another object or the moving object appears from behind the other object; when the moving object is in the shielded state, setting a time period required to complete a determination regarding the collision to a shorter time period than when the moving object is not in the shielded state; Extracting at least one shielding object positioned in at least one specific area set as at least one partial area in the captured image and suitable for shielding the moving object; and Setting at least one moving object extraction area to an area in which a field of view is determined to be shielded by the at least one shielding object positioned further in a depth direction in the captured image than the at least one specific area from which the at least one shielding object has been extracted, wherein determining whether or not the moving object is in the shielded state determines that the moving object is in the shielded state when the moving object is detected in the at least one moving object extraction area.

Images