JP6193177B2 - Motion support system and object recognition device - Google Patents

Description

  The present invention relates to a motion support system that supports motion of a moving object such as a vehicle, and an object recognition device that can be used in the motion support system.

  In patent document 1, it aims at providing the driving assistance device which can reduce unnecessary avoidance assistance ([0006], summary). In order to achieve the object, the driving support device 1 of Patent Document 1 includes an obstacle detection unit that detects information about an obstacle, and an allowance time (TTC) until the host vehicle approaches the current position of the obstacle. Margin time calculating means for calculating, future position estimating means for estimating the future position of the obstacle when the margin time has elapsed, avoidance target area setting means for setting the avoidance target area to be avoided by the host vehicle, and avoidance target Travel route setting means for setting a travel route that avoids the area (summary). The avoidance target area setting means evaluates a collision risk potential with an obstacle at a future position and its surrounding positions, and sets an avoidance target area based on the collision risk potential (summary).

  In the evaluation of the collision risk potential, the distribution of the collision risk potential at the peripheral position is generated based on the future position (S5 in FIG. 2, [0028]). That is, a pattern corresponding to the type of obstacle is acquired from the patterns stored in the risk pattern DB 4, the reference position of the pattern is matched with the future position, and the direction of the pattern is matched with the direction of the obstacle. As a result, a collision risk potential distribution around the future position is generated ([0028]). The avoidance target area is, for example, a range in which the collision risk potential is a certain threshold value or more ([0031]).

  The obstacle detection means (obstacle detection device 2) in Patent Document 1 is assumed to be a device that detects information about an obstacle using, for example, an outside camera, a radar, or the like ([0018]).

  In Patent Document 2, the obstacle detection result by radar (millimeter wave) and the obstacle detection result by image recognition are collated, and branching occurs when both are detected and only one is detected (summary). Then, by changing the start condition of the driving support control according to the branch result, the support control according to the driver's attention is executed (summary). In the case where an obstacle can be detected by each of the radar (millimeter wave) and image recognition, in Patent Document 2, if the obstacle is detected by both the millimeter wave radar 21 and the image recognition means 22, the obstacle is correctly detected. Therefore, it is described that the assist control is executed at a normal timing ([0075]).

JP 2012-173786 A JP-A-2005-239114

  In Patent Document 1 as described above, an avoidance target area is set mainly using the future position of the obstacle based on the margin time (TTC) and the type and direction of the obstacle ([0028]). In Patent Literature 1, processing in a configuration combining radar (millimeter wave) and obstacle detection results by image recognition as in Patent Literature 2 is not studied. In addition, Patent Document 1 does not consider control focusing on the reliability of the detection result of the obstacle detection means (obstacle detection device 2).

  In Patent Document 2, although the start condition of the driving support control is changed based on the obstacle detection result by each of the radar and the image recognition, the difference or change in reliability when the obstacle is detected by each of the radar and the image recognition. Has not been studied. For this reason, in Patent Document 2, it can be said that the control is performed on the premise that there is reliability when the obstacle detection result by each of the radar and the image recognition is obtained. Therefore, Patent Document 2 does not discuss the difference in reliability when the obstacle detection results obtained by the radar and the image recognition are obtained.

  The present invention has been made in consideration of the above-described problems, and an object thereof is to provide an operation support system and an object recognition apparatus that can reflect the reliability of detection results in control.

The operation support system according to the present invention includes:
A radar that emits an electromagnetic wave to a peripheral object existing around a moving object and outputs a reflected wave signal corresponding to a reflected wave from the peripheral object;
A camera that captures a peripheral image including the peripheral object and outputs an image signal corresponding to the peripheral image;
Recognizing the peripheral object included in the reflected wave signal as a first object, recognizing the peripheral object included in the image signal as a second object, and matching the first object and the second object, the target object A collision possibility determination device that specifies the collision possibility of the moving object with respect to the target object, and
An operation support apparatus for supporting the operation of the moving object according to the collision possibility;
Comprising:
A part or all of the motion support system is provided inside or outside the moving object,
The image recognition continuation time during which the recognition of the second object is continued, the lateral velocity of the second object, and information on whether the lateral velocity can be detected by either the radar or the camera. Based on the above, the possibility of collision or the motion support is changed.
The operation support system according to the present invention includes:
A radar that emits an electromagnetic wave to a peripheral object existing around a moving object and outputs a reflected wave signal corresponding to a reflected wave from the peripheral object;
A camera that captures a peripheral image including the peripheral object and outputs an image signal corresponding to the peripheral image;
Recognizing the peripheral object included in the reflected wave signal as a first object, recognizing the peripheral object included in the image signal as a second object, and matching the first object and the second object, the target object A collision possibility determination device that specifies the collision possibility of the moving object with respect to the target object, and
An operation support device for supporting the movement of the moving object according to the collision possibility,
A part or all of the motion support system is provided inside or outside the moving object,
Based on the image recognition continuation time during which the recognition of the second object continues and the lateral velocity itself or the possibility of calculation of the lateral velocity for the first object and the second object, The operation support is changed.

  When the target object is specified by combining the radar detection result (reflected wave signal) and the camera detection result (image signal), the distance to the target object can be accurately detected using the radar detection value. In many cases. In addition, when the distance to the target object calculated using the detection value of the radar is used, the peripheral image can often calculate the lateral speed of the peripheral object with higher accuracy than the detection value of the radar.

  According to the present invention, based on the image recognition continuation time during which the recognition of the second object based on the peripheral image is continued, and whether or not the lateral speed itself or the lateral speed can be calculated for the first object and the second object. Change the collision probability or motion support.

  Accordingly, by using the image recognition duration time, it is possible to reflect the reliability of the lateral position or the lateral speed based on the peripheral image in the collision possibility or the motion support. In addition, it is possible to perform operation support that more accurately reflects the reliability of the detection results of the radar and the camera by using whether or not the lateral speed itself or the lateral speed of the first object and the second object can be calculated. It becomes possible.

The collision possibility determination device includes a route overlap time calculation unit that calculates a route overlap time that is a time during which the target object is present on the route of the moving object, and the route overlap time calculation unit includes: When at least one of the current position or the future position of the target object is on the path of the moving object , a predetermined value is added to the path overlap time, and the first object is based on the image recognition duration time or In addition, the condition for adding the predetermined value to the course overlap time may be changed based on the lateral speed itself of the second object or whether the lateral speed can be calculated.

According to the present invention, the course overlap time is determined based on the image recognition duration time during which the second object is continuously recognized, or based on whether or not the lateral speed of the surrounding object itself or the lateral speed can be calculated. Change the condition for adding values .

  Thus, by using the image recognition continuation time, it is possible to reflect the reliability of the image recognition in the course overlap time. Alternatively, it is possible to calculate the course overlap time more accurately reflecting the reliability of the image recognition by using the lateral velocity of the surrounding objects or the possibility of calculating the lateral velocity.

  The collision possibility determination device may determine the collision possibility according to the course overlap time. As a result, the reliability of image recognition can be reflected in the possibility of collision.

  In performing motion support for the moving object, the motion support device may change the motion support in accordance with the course overlap time. As a result, the reliability of image recognition can be reflected in the operation support.

When the image recognition duration time is continuously recognized exceeding a predetermined time threshold, the course overlap time calculation unit, if at least the future position of the target object exists on the course of the moving object, The predetermined value may be added to the course overlap time. This makes it possible to appropriately reflect the course overlap time in a state where the reliability of image recognition is high.

When the image recognition duration time is not continuously recognized beyond the predetermined time threshold, the course overlap time calculation unit is configured such that the lateral speed of the second object is equal to or higher than a predetermined speed threshold and If the future position of the target object exists on the path of the moving object, the predetermined value may be added to the path overlap time. This makes it possible to appropriately calculate the course overlap time in a situation of high urgency where the lateral speed is relatively large.

When both the current position and the future position of the target object do not exist on the path of the moving object, the path overlap time calculation unit may subtract the predetermined value from the path overlap time. Thereby, when the possibility of collision is low, it is possible to appropriately calculate the course overlap time.

  The route overlap time calculation unit calculates the route overlap time when the current position of the target object exists on the route of the moving object and the future position of the target object does not exist on the route of the movable object. The previous value may be held. Thereby, when the possibility of collision is relatively low, it is possible to appropriately calculate the course overlap time.

When the lateral movement of both the first object and the second object is detected, the course overlap time calculation unit determines that at least the future position of the target object exists on the course of the moving object. The predetermined value may be added to the course overlap time. Thereby, when the possibility of collision is relatively high, it is possible to appropriately calculate the course overlap time.

When the first object is recognized and the second object is not recognized, the course overlap time calculation unit determines that both the current position and the future position of the target object exist on the path of the moving object. For example, the predetermined value may be added to the course overlap time. Thereby, even when the second object is not recognized, it is possible to appropriately calculate the course overlap time.

  When the first object is recognized and the second object is not recognized, the course overlap time calculation unit has only one of the current position and the future position of the target object on the course of the moving object. If so, the previous value of the course overlap time may be held. Thereby, even when the second object is not recognized, it is possible to appropriately calculate the course overlap time.

When the lateral movement of the second object is detected, the lateral movement of the first object is not detected, and the lateral speed of the second object is below a predetermined speed threshold, the path overlap time calculation unit If the current position and future position of the target object exist on the path of the moving object, the predetermined value may be added to the path overlap time. Thereby, even when the lateral movement of the first object is not detected, it is possible to appropriately calculate the course overlap time.

  When the lateral movement of the second object is detected, the lateral movement of the first object is not detected, and the lateral speed of the second object is below the predetermined speed threshold, the course overlap time calculation The unit may hold the previous value of the course overlap time if only one of the current position and the future position of the target object is present on the course of the moving object. Thereby, even when the lateral movement of the first object is not detected, it is possible to appropriately calculate the course overlap time.

An object recognition apparatus according to the present invention includes a position information acquisition apparatus that acquires position information related to positions of surrounding objects existing around a moving object, and a current position or a future position of the surrounding object based on the position information is the moving object. A route overlap time calculating unit that calculates a route overlap time that is a time existing on the route of the moving object, wherein at least one of a current position or a future position of the peripheral object is on the route of the moving object. A predetermined value is added to the route overlap time , and the position information acquisition device continues to acquire the position information of the peripheral object by the position information acquisition device, or the peripheral object changing the conditions for adding the predetermined value to the path overlap time based on whether the rate itself or calculating the speed And butterflies.

According to the present invention, the route overlap is based on the position information acquisition continuation time during which the acquisition of the position information of the peripheral object is continued, or based on whether or not the lateral speed of the peripheral object itself or the calculation of the lateral speed can be calculated. The condition for adding a predetermined value to the time is changed.

  Thus, by using the position information acquisition continuation time, the reliability of the position information can be reflected in the course overlap time. Alternatively, it is possible to calculate the course overlap time that more accurately reflects the reliability of the position information by using the speed of the surrounding object itself or whether the speed can be calculated.

  According to the present invention, the reliability of the detection result can be reflected in the control.

It is a block diagram which shows the structure of the vehicle as a driving assistance system which concerns on one Embodiment of this invention. It is a flowchart of the driving assistance control in the said embodiment. FIG. 6 is a first diagram for explaining a route overlap amount based on a current position of a target object. FIG. 6 is a second diagram for explaining the course overlap amount based on the current position of the target object. It is a figure for demonstrating the said course overlap amount based on the future position of the said target object. It is a flowchart (detail of S5 of FIG. 2) which selects the calculation mode of course overlap time. It is explanatory drawing of the selection method of the said calculation mode of the said course overlap time. It is a flowchart (detail of S6 of FIG. 2) which calculates the course overlap time. It is explanatory drawing of the calculation method of the said course overlap time.

A. Embodiment A1. Configuration [A1-1. overall structure]
FIG. 1 is a block diagram showing a configuration of a vehicle 10 (hereinafter also referred to as “own vehicle 10”) as a driving support system according to an embodiment of the present invention. The vehicle 10 includes an object recognition device 12, a vehicle behavior stabilization system 14 (hereinafter referred to as “VSA system 14”), an electric power steering system 16 (hereinafter referred to as “EPS system 16”), and a pop-up hood system 18 (hereinafter referred to as “PUH system”). 18 ”), and includes a vehicle speed sensor 20 and an alarm device 22.

  The object recognition device 12 detects various peripheral objects 100 (for example, a pedestrian 102 (FIGS. 3 and 5), another vehicle 104 (FIG. 4), and a wall) appearing around the host vehicle 10. Then, the object recognition device 12 selects or specifies a target object 100tar that is related to the control of the host vehicle 10 among the peripheral objects 100 (hereinafter also referred to as “detected object 100”). The object recognition device 12 calculates the distance L from the host vehicle 10 to the target object 100tar, and determines the attribute Prtar of the target object 100tar. The attribute Prtar here includes, for example, the type Ca (for example, vehicle, pedestrian (human) or wall) of the target object 100tar.

  An electronic control unit 30 (hereinafter referred to as “VSA ECU 30”) of the VSA system 14 performs vehicle behavior stabilization control, and the vehicle 10 is turned when turning on a curved road through control of a brake system (not shown). The behavior of the vehicle 10 is stabilized when the other vehicle 104 approaches the vehicle.

  An electronic control device 32 (hereinafter referred to as “EPS ECU 32”) of the EPS system 16 executes steering assist control, and is a constituent element of an electric power steering device {an electric motor, a torque sensor, and a steering angle sensor (all shown in FIG. Etc.) assists the steering by the driver.

  The electronic control unit 34 (hereinafter referred to as “PUH ECU 34”) of the PUH system 18 is a pop-up hood (not shown) of the vehicle 10 when the vehicle 10 encounters a collision with the pedestrian 102 (FIGS. 3 and 5). ) And the impact when the head of the pedestrian 102 hits the vehicle body is reduced.

  The vehicle speed sensor 20 detects the vehicle speed V [km / h] of the vehicle 10 and outputs it to the object recognition device 12 or the like. The warning device 22 issues a warning to the driver based on a command from the object recognition device 12 or the like. The alarm device 22 includes, for example, a display device and a speaker (not shown).

[A1-2. Object recognition device 12]
As shown in FIG. 1, the object recognition device 12 includes a radar 40, a camera 42, and an object recognition electronic control device 44 (hereinafter referred to as “object recognition ECU 44” or “ECU 44”).

(A1-2-1. Radar 40)
The radar 40 outputs a transmission wave Wt, which is an electromagnetic wave (here, a millimeter wave), to the outside of the vehicle 10, and reflects back to the detection object 100 (for example, a pedestrian 102, another vehicle 104) out of the transmission wave Wt. The incoming reflected wave Wr is received. Then, a detection signal corresponding to the reflected wave Wr (hereinafter referred to as “reflected wave signal Swr” or “signal Swr”) is output to the ECU 44. Hereinafter, the detected object 100 detected by the radar 40 is also referred to as “first object 100r” or “radar target 100r”.

  The radar 40 is disposed on the front side of the vehicle 10 (for example, a front bumper and / or a front grill). In addition to the front side or instead of the front side, the vehicle 10 may be arranged on the rear side (for example, the rear bumper and / or the rear grille) or the side (for example, the side of the front bumper).

  As will be described later, a sensor such as a laser radar or an ultrasonic sensor can be used instead of the radar 40 that outputs millimeter waves.

(A1-2-2. Camera 42)
The camera 42 (imaging means) acquires an image Imc (hereinafter also referred to as “peripheral image Imc” or “captured image Imc”) around the vehicle 10 (including the target object 100 tar). Then, a signal corresponding to the image Imc (hereinafter referred to as “image signal Sic” or “signal Sic”) is output to the ECU 44. Hereinafter, the detected object 100 detected by the camera 42 is also referred to as a “second object 100c” or a “camera target 100c”.

  In the present embodiment, one camera 42 is used, but a stereo camera may be configured by arranging two cameras 42 symmetrically. The camera 42 acquires the image Imc at 15 frames or more (for example, 30 frames) per second. The camera 42 is a monochrome camera that mainly uses light having a wavelength in the visible light region, but may be a color camera or an infrared camera. The camera 42 is disposed, for example, at the center in the vehicle width direction (for example, around the rearview mirror) in the front portion of the vehicle interior of the vehicle 10. Alternatively, the camera 42 may be disposed at the center in the vehicle width direction in the front bumper portion of the vehicle 10.

(A1-2-3. Object recognition ECU 44)
The object recognition ECU 44 controls the entire object recognition device 12, and includes an input / output unit 50, a calculation unit 52, and a storage unit 54 as shown in FIG.

  The reflected wave signal Swr from the radar 40 and the image signal Sic from the camera 42 are supplied to the object recognition ECU 44 via the input / output unit 50. Communication between the object recognition ECU 44 and the VSA ECU 30, EPS ECU 32, and PUH ECU 34 is performed via the input / output unit 50 and the communication line 56. The input / output unit 50 includes an A / D conversion circuit (not shown) that converts an input analog signal into a digital signal.

  The calculation unit 52 performs calculations based on the signals Swr and Sic from the radar 40 and the camera 42, and generates signals for the VSA ECU 30, EPS ECU 32, and PUH ECU 34 based on the calculation results.

  As illustrated in FIG. 1, the calculation unit 52 includes a radar information processing unit 60, a camera information processing unit 62, and a target object information processing unit 64. Each processing unit 60, 62, 64 is realized by executing a program stored in the storage unit 54. The program may be supplied from the outside via a wireless communication device (mobile phone, smartphone, etc.) not shown. A part of the program can be configured by hardware (circuit parts).

  The radar information processing unit 60 detects information (hereinafter “radar information Ir” and “first object information Ir”) of the detected object 100 (first object 100r) based on the reflected wave Wr (reflected wave signal Swr) detected by the radar 40. Or “information Ir”). The camera information processing unit 62 uses the peripheral image Imc acquired by the camera 42 to provide information on the detected object 100 (second object 100c) (hereinafter referred to as “camera information Ic”, “second object information Ic”, or “information Ic”). .) Is calculated.

  The target object information processing unit 64 (hereinafter also referred to as “processing unit 64”) is information (a combination of the radar information Ir calculated by the radar information processing unit 60 and the camera information Ic calculated by the camera information processing unit 62). Hereinafter, “target object information It” or “information It” is calculated. In other words, the processing unit 64 performs so-called fusion processing. The information It is information on the target object 100tar specified based on the detected object 100 (first object 100r) detected by the radar 40 and the detected object 100 (second object 100c) detected by the camera 42.

  The processing unit 64 collides with a TTC calculation unit 70 (hereinafter also referred to as “calculation unit 70”) and a course overlap time calculation unit 72 (hereinafter also referred to as “Tolp calculation unit 72” or “calculation unit 72”). Possibility calculation unit 74 (hereinafter also referred to as “Pc calculation unit 74” or “calculation unit 74”).

  The TTC calculation unit 70 calculates TTC (Time to Collision) indicating the time until the host vehicle 10 contacts (or collides) with the target object 100tar. The course overlap time calculation unit 72 calculates a course overlap time Tolp. The time Tolp is a time during which the target object 100tar is present on the route 106 (FIGS. 3 to 5) of the vehicle 10. Details of the time Tolp will be described later with reference to FIGS.

  The collision possibility calculation unit 74 calculates the collision possibility Pc. The collision possibility Pc is a possibility that the target object 100tar (pedestrian 102 or the like) contacts (or collides) with the own vehicle 10.

  The storage unit 54 includes an imaging signal converted into a digital signal, a RAM (Random Access Memory) that stores temporary data used for various arithmetic processes, and a ROM (Read Only Memory) that stores an execution program, a table, a map, or the like. ) Etc.

A2. Driving support control [A2-1. Overall flow of driving support control]
FIG. 2 is a flowchart of the driving support control in the present embodiment. The driving support control is executed by the calculation unit 52 (processing units 60, 62, 64) of the object recognition ECU 44 and the calculation units (not shown) of the other ECUs 30, 32, 34. Each ECU 30, 32, 34, 44 repeats the process shown in FIG. 2 at a predetermined calculation cycle (for example, any cycle of several μsec to several hundred msec).

  In step S1, the ECU 44 (radar information processing unit 60) calculates radar information Ir of the detection object 100 (radar target 100r) based on the reflected wave signal Swr from the radar 40. The radar information Ir includes the position Por of the radar target 100r, the velocity Vr, the acceleration ar, and the like. If there are a plurality of radar targets 100r within the detection range of the radar 40, the ECU 44 calculates the position Por, the velocity Vr, the acceleration ar, etc. for each radar target 100r.

  In step S2, the ECU 44 (camera information processing unit 62) calculates camera information Ic of the detected object 100 (camera target 100c) based on the image signal Sic (captured image Imc) of the camera 42. The camera information Ic includes the position Poc of the camera target 100c, the lateral speed Vlc, the acceleration ac, the attribute Prc, and the like. Further, the attribute Prc includes the type Ca (pedestrian, vehicle, etc.), size, etc. of the camera target 100c. When there are a plurality of camera targets 100c in the captured image Imc, the ECU 44 calculates the position Poc, the lateral speed Vlc, the acceleration ac, the attribute Prc, and the like for each camera target 100c.

  In step S3, the ECU 44 (target object information processing unit 64) performs a matching determination for matching the radar information Ir (radar target 100r) and the camera information Ic (camera target 100c). The matching determination is the same when the position Por of the first object 100r recognized by the radar information processing unit 60 and the position Poc of the second object 100c recognized by the camera information processing unit 62 match or are within a predetermined distance. It is determined that the target object is 100 tar.

  In step S4, the ECU 44 calculates TTC indicating the time until the host vehicle 10 contacts (or collides with) the target object 100tar.

  In step S5, the ECU 44 selects a calculation mode Molp (hereinafter also referred to as “mode Molp”) for the course overlap time Tolp (hereinafter also referred to as “overlap time Tolp” or “time Tolp”). The time Tolp is a time during which the target object 100tar is present on the route 106 (FIGS. 3 to 5) of the vehicle 10. The calculation mode Molp is a mode for calculating the time Tolp, and in the present embodiment, Mode 1 and Mode 2 are used. Details of step S5 will be described later with reference to FIG.

  In step S6, the ECU 44 calculates the overlap time Tolp using the calculation mode Molp selected in step S5. Details of step S6 will be described later with reference to FIG.

  In step S7, the ECU 44 calculates a collision possibility Pc from the TTC and the overlap time Tolp. The collision possibility Pc is a possibility that the target object 100tar (pedestrian 102 or the like) contacts (or collides) with the own vehicle 10. The calculated collision possibility Pc is output from the ECU 44 to the ECUs 30, 32, and 34, for example.

  It is also possible to use the TTC and the time Tolp as they are without using the collision possibility Pc. In that case, for example, an alarm using the alarm device 22 or an automatic brake using an unillustrated brake device can be performed in relation to the threshold values of the TTC and the time Tolp.

  Further, it is possible to determine the collision possibility Pc only when the time Tolp exceeds the threshold value, or to change the TTC threshold value according to the time Tolp.

  In step S8, the ECU 44 performs driving assistance according to the collision possibility Pc. Specifically, the VSA ECU 30 executes automatic brake control using, for example, the collision possibility Pc. The EPS ECU 32 executes automatic steering control using, for example, the collision possibility Pc. The PUH ECU 34 executes PUH control using the collision possibility Pc. The object recognition ECU 44 issues a predetermined warning (for example, output of warning sound or warning display) via the alarm device 22.

[A2-2. Course overlap time Tolp]
(A2-2-1. Significance of route overlap time Tolp)
As described above, in the present embodiment, the collision possibility Pc is calculated using the course overlap time Tolp in addition to the TTC. That is, the time Tolp is used as an index for determining the collision possibility Pc. When the time Tolp value is small, it is determined that the collision possibility Pc is relatively low, and when the time Tolp value is large, it is determined that the collision possibility Pc is relatively high.

(A2-2-2. Outline of calculation of route overlap time Tolp)
In this embodiment, after selecting the calculation mode Molp for the course overlap time Tolp (S5 in FIG. 2), the time Tolp is calculated. The calculation mode Molp includes Mode 1 and Mode 2. Mode 1 is used when the reliability (accuracy) of the lateral velocity Vlc of the second object 100c by image recognition is relatively high. Mode 2 is used when the reliability (accuracy) of the lateral velocity Vlc of the second object 100c by image recognition is relatively low. Details of each mode Molp will be described in detail with reference to FIGS.

  Regardless of which calculation mode Molp is used, the presence or absence of a route overlap is used (FIG. 9). First, the route overlap will be described.

  The route overlap means a state in which the route overlap amount Dolp is equal to or greater than a predetermined threshold THdolp (hereinafter also referred to as “overlap amount threshold THdolp”). In the present embodiment, the presence or absence of a route overlap is determined for each of the current position Ppre and the future position Pftr of the target object 100tar.

(A2-2-3. Route overlap amount Dolp)
(A2-2-3-1. Course overlap amount Dolp based on the current position Ppre)
FIGS. 3 and 4 are FIGS. 1 and 2 for explaining the route overlap amount Dolp based on the current position Ppre of the target object 100tar. In FIG. 3, a pedestrian 102 as the target object 100 tar exists in front of the host vehicle 10. In FIG. 4, another vehicle 104 as the target object 100 tar is present in front of the host vehicle 10. 3 and 4, the broken line 108 l is a virtual line indicating the left end of the route 106 of the vehicle 10, and the broken line 108 r is a virtual line indicating the right end of the route 106 of the vehicle 10. A region sandwiched between the virtual lines 108l and 108r is the course 106.

  3 and 4, the broken lines 108l and 108r correspond to the position of the side surface of the vehicle 10. However, for the purpose of calculating the course 106 of the vehicle 10, it can be displaced in the vehicle width direction. .

  3 and 4, the alternate long and short dash line 110l is an imaginary line indicating the left end of the target object 100tar (pedestrian 102, other vehicle 104) when viewed from the vehicle 10. The alternate long and short dash line 110r is a virtual line indicating the right end of the target object 100tar (pedestrian 102, other vehicle 104) when viewed from the vehicle 10. Like the broken lines 108l and 108r, the alternate long and short dash lines 110l and 110r can be displaced in the vehicle width direction.

  In FIGS. 3 and 4, the virtual lines 108 l and 108 r are straight lines. However, for example, when the vehicle 10 is turning, the virtual lines 108 l and 108 r may be curved according to the course 106. In order to make the virtual lines 108l and 108r curved, for example, the virtual lines 108l and 108r can be calculated using a yaw rate detected by a yaw rate sensor (not shown). Further, information on the shape of the route 106 is acquired from a navigation device (not shown) or a device (roadside device) such as an optical beacon arranged around the route 106 (traveling route), and the virtual lines 108l and 108r are obtained using the information. It is also possible to calculate.

  3 and 4, Dolp_l is the overlap amount of the target object 100tar (pedestrian 102 or other vehicle 104) with respect to the course 106 when the left virtual line 108l is used as a reference. That is, the overlap amount Dolp_l indicates a distance from a virtual line 108l indicating the left end of the route 106 to a virtual line 110r indicating the right end of the target object 100tar (pedestrian 102 or other vehicle 104). However, in FIG. 4, the virtual line 110 r indicating the right end of the other vehicle 104 is located on the right side of the virtual line 108 r indicating the right end of the route 106 and does not exist in the route 106. Therefore, in the example of FIG. 4, the overlap amount Dolp_l is the distance between the virtual lines 108l and 108r, and is the maximum value that the overlap amount Dolp_l can take.

  Similarly, Dolp_r is the overlap amount of the target object 100tar (pedestrian 102 or other vehicle 104) with respect to the course 106 when the right virtual line 108r is used as a reference. That is, the overlap amount Dolp_r indicates a distance from a virtual line 108r indicating the right end of the route 106 to a virtual line 110l indicating the left end of the target object 100tar. However, in FIG. 4, the virtual line 110 l indicating the left end of the other vehicle 104 is located on the left side of the virtual line 108 l indicating the left end of the route 106 and does not exist in the route 106. Therefore, in the example of FIG. 4, the overlap amount Dolp_r is the distance between the virtual lines 108l and 108r, and is the maximum value that the overlap amount Dolp_r can take.

  3 and 4 show overlap amounts Dolp_r and Dolp_r based on the current position Ppre of the target object 100tar.

  The ECU 44 uses the shorter one of the route overlap amounts Dolp_l and Dolp_r as the overlap amount Dolp. When the overlap amounts Dolp_l and Dolp_r are equal to each other, any value may be used. In the example of FIG. 3, the left route overlap amount Dolp_l is used as the route overlap amount Dolp_. In the example of FIG. 4, since the left and right route overlap amounts Dolp_l and Dolp_r are equal, one of the route overlap amounts Dolp_l and Dolp_r (for example, Dolp_l) is used as the route overlap amount Dolp. In the example of FIGS. 3 and 4, the overlap amount Dolp based on the current position Ppre of the target object 100tar is calculated.

(A2-2-3-2. Course overlap amount Dolp based on future position Pftr)
FIG. 5 is a diagram for explaining the route overlap amount Dolp based on the future position Pftr of the target object 100tar. FIG. 5 shows a state where a pedestrian 102 as the target object 100 tar enters the course 106 of the host vehicle 10. In FIG. 5, the vehicle 10 and the pedestrian 102 at the current position Ppre are indicated by solid lines, and the vehicle 10 and the pedestrian 102 at the future position Pftr are indicated by broken lines. The future position Pftr here is calculated based on, for example, TTC.

  Similar to the overlap amount Dolp based on the current position Ppre, the overlap amount Dolp based on the future position Pftr calculates the left and right overlap amounts Dolp_l and Dolp_r, and sets the smaller value as the overlap amount Dolp.

(A2-2-4. Course overlap time Tolp calculation mode Molp)
(A2-2-4-1. Overview of calculation mode Molp)
As described above, the calculation mode Molp of the present embodiment includes Mode 1 and Mode 2 corresponding to the reliability of the lateral velocity Vlc of the second object 100c by image recognition.

(A2-2-4-2. Selection of calculation mode Molp)
FIG. 6 is a flowchart (details of S5 in FIG. 2) for selecting the calculation mode Molp for the route overlap time Tolp. FIG. 7 is an explanatory diagram of a method for selecting the calculation mode Molp for the course overlap time Tolp.

  In step S11 of FIG. 6, the ECU 44 determines whether or not object recognition is impossible for at least the reflected wave Wr. More specifically, the ECU 44 determines whether or not the first object 100r can be recognized based on the reflected wave signal Swr.

  When the object recognition is not possible for the reflected wave Wr (S11: YES), the ECU 44 determines that the calculation mode Molp cannot be selected in step S12. In this case, in step S6 of FIG. 2, the ECU 44 does not calculate the course overlap time Tolp and sets the time Tolp to zero.

  When object recognition is possible for at least the reflected wave Wr (S11: NO), in step S13, the ECU 44 determines whether object recognition is possible for both the reflected wave Wr and the image Imc. If object recognition is possible for both the reflected wave Wr and the image Imc (S13: YES), the process proceeds to step S14.

  Note that the fact that the first object 100r can be recognized by the reflected wave Wr in steps S11 and S13 means that the presence of the first object 100r can be recognized. Even if the presence of the first object 100r can be recognized, there is a case where the lateral movement of the first object 100r cannot be detected (see S16 described later). This is because fluctuation occurs in the reflected wave Wr received by the radar 40. That is, since the radar information processing unit 60 specifies the position Por of the first object 100r in consideration of fluctuations in the reflected wave Wr, there may occur a case where the velocity Vr cannot be detected even if the position Por can be detected.

  In step S14, the ECU 44 determines whether or not the duration time Tcr of the image recognizable state (hereinafter also referred to as “image recognition duration time Tcr”) is sufficiently long. Specifically, it is determined whether or not the duration Tcr is equal to or greater than a threshold value THtcr (hereinafter also referred to as “duration threshold value THtcr”).

  Here, “the duration Tcr is sufficiently long” means that the image recognizable state continues so as to ensure that the lateral velocity Vlc of the second object 100c by image recognition can be accurately detected. . That is, when the duration time Tcr is short, the number of data of the image Imc for calculating the lateral speed Vlc is small, and therefore there is an error (or a sudden erroneous recognition) in the lateral speed Vlc of the second object 100c due to image recognition. Is more likely to occur. On the other hand, when the duration Tcr is increased, the reliability (accuracy) of the lateral speed Vlc of the second object 100c by image recognition is increased. Therefore, as in step S14, the process is changed according to the degree of reliability of the lateral speed Vlc based on the image Imc.

  If the duration Tcr is not sufficiently long (S14 in FIG. 6: NO), for example, the reliability is ensured by using the lateral velocity Vlr of the first object 100r and the lateral velocity Vlc of the second object 100c. (S16 in FIG. 6 described later: YES, FIG. 7). When the duration Tcr is sufficiently long (S14: YES), only the lateral speed Vlc of the second object 100c is used instead of the lateral speed Vlr of the first object 100r in calculating the overlap amount Dolp. This improves reliability (Fig. 7).

  When the duration time Tcr is not sufficiently long (S14: NO), the reliability of the lateral speed Vlc of the camera target 100c is relatively low. In this case, in step S15, the ECU 44 determines whether or not the lateral speed Vlc of the camera target 100c is sufficiently large. Specifically, it is determined whether or not the lateral speed Vlc is equal to or higher than a threshold value THvlc (hereinafter also referred to as “lateral speed threshold value THvlc”).

  Here, “the lateral speed Vlc is sufficiently large” means that, for example, even if the duration Tcr is short and the reliability of the lateral speed Vlc is low, the lateral speed Vlc is relatively high, so the duration Tcr Means that the lateral velocity Vlc is so large that it cannot wait for the duration to continue. From another point of view, the threshold value THvlc can be set to a value (for example, 4 to 20 km / h) that clearly indicates the pedestrian 102 that is walking or running or the other vehicle 104 that is running.

  If the lateral speed Vlc is not sufficiently high (S15: NO), in step S16, the ECU 44 determines whether or not the lateral movement of the radar target 100r and the camera target 100c can be detected. In other words, the ECU 44 determines whether or not lateral movement can be detected by the reflected wave Wr and the image Imc. As described above, even if the presence of the radar target 100r can be recognized, the lateral movement of the radar target 100r may not be detected. In step S16, determination based on this is performed.

  If any of steps S14, S15, and S16 is “YES”, in step S17, the ECU 44 selects Mode 1 as the calculation mode Molp.

  If “NO” in step S13 or if all of steps S14, S15, and S16 are “NO”, in step S18, the ECU 44 selects Mode 2 as the calculation mode Molp.

(A2-2-5. Calculation of route overlap time Tolp)
(A2-2-5-1. Overall flow)
As described above, in the present embodiment, after selecting the calculation mode Molp, the route overlap time Tolp is calculated based on the presence or absence of the route overlap at the current position Ppre and the future position Pftr (FIG. 2 and the like).

  FIG. 8 is a flowchart (details of S6 in FIG. 2) for calculating the course overlap time Tolp. FIG. 9 is an explanatory diagram of a method for calculating the course overlap time Tolp.

  In step S21 of FIG. 8, the ECU 44 determines whether or not there is a course overlap at the current position Ppre. Specifically, the ECU 44 determines that there is a course overlap at the current position Ppre if the overlap amount Dolp (FIGS. 3 and 4) based on the current position Ppre is equal to or greater than a predetermined threshold value THdolp. If the overlap amount Dolp based on the current position Ppre is less than the threshold THdolp, the ECU 44 determines that there is no course overlap at the current position Ppre.

  In step S22, the ECU 44 determines whether or not there is a route overlap at the future position Pftr. Specifically, the ECU 44 determines that there is a course overlap at the future position Pftr if the overlap amount Dolp (FIG. 5) based on the future position Pftr is greater than or equal to a predetermined threshold value THdolp. In addition, if the overlap amount Dolp based on the future position Pftr is less than the threshold value THdolp, the ECU 44 determines that there is no course overlap at the future position Pftr. Note that the threshold value THdolp may be changed between the current position Ppre and the future position Pftr.

  In step S23, the ECU 44 calculates a course overlap time Tolp based on the determination results of steps S21 and S22. At this time, the ECU 44 uses the calculation mode Molp selected in step S5 (FIG. 7) in FIG. Details of the method of calculating the time Tolp will be described with reference to FIG.

(A2-2-5-2. Relationship between presence / absence of route overlap and route overlap time Tolp)
FIG. 9 is a diagram illustrating the relationship between the presence / absence of the route overlap at the current position Ppre and the future position Pftr and the addition or subtraction of the route overlap time Tolp for each calculation mode Molp.

  When there is a course overlap in both the current position Ppre and the future position Pftr, the ECU 44 adds the predetermined value (here, 1) to the previous value of the time Tolp and increases the time Tolp regardless of the calculation mode Molp. The case where there is a course overlap at both the current position Ppre and the future position Pftr indicates a state in which the lateral speed Vlt of the target object 100tar (or the lateral speed Vlc of the camera target 100c) is relatively low. Yes.

  When there is no course overlap at both the current position Ppre and the future position Pftr, the ECU 44 subtracts a predetermined value (here, 1) from the previous value of the time Tolp and decreases the time Tolp regardless of the calculation mode Molp.

  When there is a course overlap at the current position Ppre and no course overlap at the future position Pftr, the ECU 44 maintains the previous value of the time Tolp and does not change the time Tolp regardless of the calculation mode Molp.

  When there is no course overlap at the current position Ppre and there is a course overlap at the future position Pftr, the ECU 44 calculates the time Tolp according to the calculation mode Molp.

  That is, when the calculation mode Molp is Mode 1, the ECU 44 adds the predetermined value (here, 1) to the previous value of the time Tolp and increases the time Tolp. Mode 1 is because the reliability of the lateral velocity Vlc is relatively high, and therefore it can be evaluated that the collision possibility Pc is high.

  On the other hand, when the calculation mode Molp is Mode 2, the ECU 44 keeps the previous value of the time Tolp and does not change the time Tolp. Mode 2 is because the reliability of the lateral velocity Vlc is relatively low, and therefore cannot be evaluated when the collision possibility Pc is high.

  As can be seen from the above, the time Tolp is calculated in common with the route overlap at the current position Ppre and the future position Pftr. In other words, in the present embodiment, the time Tolp corresponding to the current position Ppre and the time Tolp corresponding to the future position Pftr are not set separately. However, the time Tolp corresponding to the current position Ppre and the time Tolp corresponding to the future position Pftr can be set separately, weighted and added.

  Further, the amount of increase in the time Tolp in one calculation cycle is constant. For this reason, for example, even when there is a course overlap at both the current position Ppre and the future position Pftr, the increase amount does not increase. However, it is possible to accompany such a change in the amount of increase.

  Further, the lap amount Dolp is used to determine whether or not there is a route overlap, and does not affect the increase amount of the time Tolp in one calculation cycle. In other words, the increase amount of the time Tolp in one calculation cycle is not affected by the magnitude of the lap amount Dolp. For example, even if the lap amount Dolp is the maximum value, the increase amount does not increase. However, it is possible to accompany such a change in the amount of increase.

A3. As described above, according to the present embodiment, the image recognition continuation time Tcr in which the recognition of the second object 100c based on the peripheral image Imc is continued, the first object 100r, and the second object 100c. The possibility of collision Pc or driving assistance is changed based on whether or not the lateral speeds Vlr and Vlc themselves or the lateral speeds Vlr and Vlc can be calculated (FIGS. 2 and 6 to 9).

  Thus, by using the image recognition duration time Tcr, the reliability of the lateral position or the lateral speed Vlc based on the peripheral image Imc can be reflected in the collision possibility Pc or the driving assistance. In addition, the reliability of the detection results of the radar 40 and the camera 42 can be further improved by using the possibility of calculating the lateral velocities Vlr and Vlc themselves or the lateral velocities Vlr and Vlc for the first object 100r and the second object 100c. It is possible to provide driving assistance that accurately reflects this.

  In the present embodiment, the target object information processing unit 64 (collision possibility determination device) calculates the course overlap time Tolp, which is the time during which the target object 100tar exists on the course 106 of the vehicle 10 (moving object). A lap time calculation unit 72 is provided (FIG. 1). When both the current position Ppre and future position Pftr of the target object 100tar or only the future position Pftr exists on the course 106, the calculation unit 72 adds the course overlap time Tolp (FIG. 9). The calculation unit 72 determines the course overlap time Tolp based on the duration time Tcr or based on whether or not the lateral speeds Vlr and Vlc themselves or the lateral speeds Vlr and Vlc can be calculated for the first object 100r and the second object 100c. Is changed (FIGS. 6, 7 and 9).

  According to the present embodiment, based on the image recognition continuation time Tcr in which the recognition of the second object 100c is continued (or the position information acquisition continuation time in which the acquisition of the position information of the peripheral object 100 is continued), or around The conditions for adding the course overlap time Tolp are changed based on whether the lateral speeds Vlr and Vlc themselves or the lateral speeds Vlr and Vlc can be calculated for the object 100 (FIGS. 6, 7 and 9).

  Thereby, by using the continuation time Tcr, it is possible to reflect the reliability of image recognition (position information) in the overlap time Tolp. In addition, the overlap time Tolp that more accurately reflects the reliability of the image recognition (position information) is calculated by using whether or not the lateral velocity Vlc itself or the lateral velocity Vlc can be calculated for the peripheral object 100. Is possible.

  In the present embodiment, the target object information processing unit 64 (collision possibility determination device) determines the collision possibility Pc according to the course overlap time Tolp (S7 in FIG. 2). As a result, the reliability of image recognition (position information) can be reflected in the collision possibility Pc.

  In the present embodiment, when performing driving support for the vehicle 10 (moving object), the object recognition device 12, the VSA system 14, the EPS system 16, and the PUH system 18 (motion support device) correspond to the route overlap time Tolp. The driving support is changed (S8 in FIG. 2). This makes it possible to reflect the reliability of image recognition (position information) in driving support.

  In the present embodiment, when the image recognition duration Tcr is continuously recognized exceeding the duration threshold THtcr (time threshold) (S14 in FIG. 6: YES), the course overlap time calculator 72 is at least the target If the future position Pftr of the object 100tar exists on the route 106 of the vehicle 10, the route overlap time Tolp is added (FIG. 9). This makes it possible to appropriately reflect the course overlap time Tolp in a state where the reliability of image recognition (position information) is high.

  In the present embodiment, when the image recognition duration Tcr exceeds the duration threshold THtcr (time threshold) and is not continuously recognized (S14 in FIG. 6: NO), the course overlap time calculation unit 72 is a camera object. If the lateral velocity Vlc of the target 100c (second object 100c) is equal to or greater than the predetermined velocity threshold THvlc (S15: YES) and at least the future position Pftr of the target object 100tar exists on the route 106, the route overlap time Tolp Are added (FIG. 9). As a result, the route overlap time Tolp can be appropriately calculated in a highly urgent situation where the lateral speed Vlc is relatively large.

  In the present embodiment, when both the current position Ppre and the future position Pftr of the target object 100tar do not exist on the route 106 of the vehicle 10, the route overlap time calculation unit 72 subtracts the route overlap time Tolp (FIG. 9). Thereby, when the collision possibility Pc is low, it is possible to appropriately calculate the course overlap time Tolp.

  In the present embodiment, the route overlap time calculation unit 72 determines the route overlap when the current position Ppre of the target object 100tar exists on the route 106 of the vehicle 10 and the future position Pftr of the target object 100tar does not exist on the route 106. The previous value of the lap time Tolp is held (FIG. 9). As a result, when the collision possibility Pc is relatively low, it is possible to appropriately calculate the course overlap time Tolp.

  In this embodiment, when the lateral movement of both the first object 100r and the second object 100c is detected (S16 in FIG. 6: YES), the course overlap time calculation unit 72 at least the future of the target object 100tar. If the position Pftr exists on the route 106 of the vehicle 10, the route overlap time Tolp is added (FIG. 9). Thereby, when the collision possibility Pc is relatively high, it is possible to appropriately calculate the course overlap time Tolp.

  In the present embodiment, when the first object 100r is recognized and the second object 100c is not recognized (S13: NO in FIG. 6), the course overlap time calculation unit 72 determines the current position Ppre and the future of the target object 100tar. If both of the positions Pftr exist on the route 106 of the vehicle 10, the route overlap time Tolp is added (FIG. 9). Thereby, even when the second object 100c is not recognized, the course overlap time Tolp can be appropriately calculated.

  In the present embodiment, when the first object 100r is recognized and the second object 100c is not recognized (S13: NO in FIG. 6), the course overlap time calculation unit 72 determines the current position Ppre and the future of the target object 100tar. If only one of the positions Pftr exists on the route 106 of the vehicle 10, the previous value of the route overlap time Tolp is held (FIG. 9). Thereby, even when the second object 100c is not recognized, the course overlap time Tolp can be appropriately calculated.

  The lateral movement of the second object 100c is detected, the lateral movement of the first object 100r is not detected (S16 in FIG. 6: NO), and the lateral speed Vlc of the second object 100c is equal to the lateral speed threshold THvlc ( (S15: NO), the route overlap time calculation unit 72 determines that the route overlap time Tolp is present if the current position Ppre and the future position Pftr of the target object 100tar are present on the route 106 of the vehicle 10. Are added (FIG. 9). Thereby, even when the lateral movement of the first object 100r is not detected, it is possible to appropriately calculate the course overlap time Tolp.

  In this embodiment, the lateral movement of the second object 100c is detected, the lateral movement of the first object 100r is not detected (S16: NO in FIG. 6), and the lateral speed Vlc of the second object 100c is horizontal. When the direction speed threshold THVlc (speed threshold) is not reached (S15: NO), the course overlap time calculation unit 72 has only one of the current position Ppre and the future position Pftr of the target object 100tar on the course 106 of the vehicle 10. Then, the previous value of the course overlap time Tolp is held (FIG. 9). Thereby, even when the lateral movement of the first object 100r is not detected, it is possible to appropriately calculate the course overlap time Tolp.

B. Modifications It should be noted that the present invention is not limited to the above-described embodiment, and it is needless to say that various configurations can be adopted based on the description of the present specification. For example, the following configuration can be adopted.

B1. Application target In the above-described embodiment, the object recognition device 12 is applied to the vehicle 10, but the present invention is not limited thereto, and may be applied to another target. For example, the object recognition device 12 can be used for a moving object such as a ship or an aircraft. Alternatively, the object recognition device 12 may be applied to a robot, a security monitoring device, or a home appliance. Further, the object recognition device 12 may be arranged outside the vehicle 10 (for example, a roadside device such as an optical beacon) instead of being mounted on the vehicle 10 (moving object) itself. In this case, it is also possible to communicate between the vehicle 10 and the object recognition device 12 and transmit the recognition result (collision possibility Pc or target object information It) of the object recognition device 12 to the vehicle 10.

B2. Configuration of the Object Recognition Device 12 In the above embodiment, the output of the object recognition device 12 (the collision possibility Pc or the target object information It) is used by the object recognition device 12, the VSA ECU 30, the EPS ECU 32, and the PUH ECU 34 (FIG. 2). S8), and can be used for other purposes. For example, it can be used for parking assistance of the vehicle 10.

  In the above embodiment, the radar 40 using the transmission wave Wt and the reflection wave Wr that are millimeter waves is used. For example, the information Ir of the first object 100r is acquired using the reflection wave Wr of the transmission wave Wt as an electromagnetic wave. From this point of view, a sensor such as a laser radar or an ultrasonic sensor can be used.

  In the above embodiment, the radar 40 and the camera 42 are used in combination (FIG. 1). However, for example, from the viewpoint of using the image recognition duration Tcr (S14 in FIG. 6), the present invention can be applied to a configuration using only the camera 42 (image Imc). On the contrary, the present invention may be applied to a configuration using only the radar 40 (reflected wave Wr) by setting an index similar to the duration Tcr for the radar 40 (reflected wave Wr).

B3. Control of the object recognition ECU 44 [B3-1. Selection of calculation mode Molp]
In the above embodiment, the calculation mode Molp is selected by the method shown in FIGS. However, it is not limited to this from the viewpoint of selecting the calculation mode Molp. For example, from the viewpoint of the reliability of the lateral velocity Vlc of the second object 100c, the calculation modes Molp can be classified into three or more types.

  In the above embodiment, the calculation mode Molp is calculated based on the lateral velocity Vlc of the second object 100c, but the velocity of the second object 100c in other directions can also be used. Also, the calculation mode Molp can be selected using the lateral speed Vlr of the first object 100r or other speeds instead of the second object 100c. For example, since the radar 40 has higher detection accuracy in the front-rear direction than the camera 42, the radar 40 may select the calculation mode Molp based on the front-rear speed of the first object 100r.

[B3-2. Calculation of route overlap time Tolp]
In the above embodiment, the time Tolp is calculated by the method shown in FIGS. However, it is not limited to this from the viewpoint of calculating the time Tolp. For example, the time Tolp can be calculated using only one of the current position Ppre and the future position Pftr. Further, the overlap time Tolp can be calculated by changing a part of FIG.

  In the above embodiment, when it is recognized that the reliability of the image Imc is high, the predetermined value (1) is added to the time Tolp, and when it is recognized that the reliability is low, the predetermined value (1) is subtracted from the time Tolp ( FIG. 9). However, from the viewpoint of changing the time Tolp according to the reliability of the image Imc, the present invention is not limited to this. For example, when the reliability of the image Imc is recognized to be high, a predetermined value to be added to the time Tolp may be set to -1, and when the reliability is recognized to be low, the predetermined value to be subtracted from the time Tolp may be set to -1.

10 ... Vehicle (moving object, motion support system)
12 ... Object recognition device (motion support device)
14 ... VSA system (operation support device)
16 ... EPS system (operation support device)
18 ... PUH system (operation support device)
40. Radar (positional information acquisition device)
42. Camera (position information acquisition device)
64 ... Target object information processing unit (collision possibility determination device)
72 ... Path overlap time calculation unit 100 ... Peripheral object 100c ... Second object 100r ... First object 100tar ... Target object 106 ... Path Imc ... Peripheral image Pc ... Collision possibility Sic ... Image signal Swr ... Reflected wave signal Tcr ... Image Recognition duration THtcr: duration threshold (time threshold)
THvlc: Lateral speed threshold (speed threshold)
Tolp ... Route overlap time Vlc ... Lateral speed Vlr of the second object ... Lateral speed of the first object Wr ... Reflected wave Wt ... Transmitted wave (electromagnetic wave)

Claims (14)

A radar that emits an electromagnetic wave to a peripheral object existing around a moving object and outputs a reflected wave signal corresponding to a reflected wave from the peripheral object;
A camera that captures a peripheral image including the peripheral object and outputs an image signal corresponding to the peripheral image;
Recognizing the peripheral object included in the reflected wave signal as a first object, recognizing the peripheral object included in the image signal as a second object, and matching the first object and the second object, the target object A collision possibility determination device that specifies the collision possibility of the moving object with respect to the target object, and
A motion support system comprising: a motion support device that supports the motion of the moving object according to the collision possibility,
A part or all of the motion support system is provided inside or outside the moving object,
An image recognition duration recognition of the second object is continued, pre-SL and lateral velocity of the second object, the said radar and information on whether or not able to detect the lateral velocity at the camera The motion support system, wherein the collision possibility or the motion support is changed based on. A radar that emits an electromagnetic wave to a peripheral object existing around a moving object and outputs a reflected wave signal corresponding to a reflected wave from the peripheral object;
A camera that captures a peripheral image including the peripheral object and outputs an image signal corresponding to the peripheral image;
Recognizing the peripheral object included in the reflected wave signal as a first object, recognizing the peripheral object included in the image signal as a second object, and matching the first object and the second object, the target object A collision possibility determination device that specifies the collision possibility of the moving object with respect to the target object, and
An operation support apparatus for supporting the operation of the moving object according to the collision possibility;
An operation support system comprising:
A part or all of the motion support system is provided inside or outside the moving object,
Based on the image recognition continuation time during which the recognition of the second object continues and the lateral velocity itself or the possibility of calculation of the lateral velocity for the first object and the second object, Changing the operation support,
The collision possibility determination device further includes a route overlap time calculation unit that calculates a route overlap time that is a time during which the target object exists on the route of the moving object,
The route overlap time calculation unit
When at least one of the current position or the future position of the target object exists on the path of the moving object , a predetermined value is added to the path overlap time,
Conditions for adding the predetermined value to the course overlap time based on the image recognition continuation time, or based on whether or not the lateral velocity of the first object and the second object itself or the lateral velocity can be calculated An action support system characterized by changing The operation support system according to claim 2,
The said collision possibility determination apparatus determines the said collision possibility according to the said course overlap time. The operation | movement assistance system characterized by the above-mentioned. The operation support system according to claim 2 or 3,
In performing motion support for the moving object, the motion support device changes the motion support according to the course overlap time. In the operation support system according to any one of claims 2 to 4,
When the image recognition duration time is continuously recognized exceeding a predetermined time threshold, the course overlap time calculation unit, if at least the future position of the target object exists on the course of the moving object, The operation support system , wherein the predetermined value is added to the course overlap time. The operation support system according to any one of claims 2 to 5,
When the image recognition duration time is not continuously recognized exceeding a predetermined time threshold, the course overlap time calculation unit is configured such that a lateral speed of the second object is equal to or greater than a predetermined speed threshold and at least the If the future position of the target object exists on the path of the moving object, the predetermined value is added to the path overlap time. In the operation support system according to any one of claims 2 to 6,
When both the current position and the future position of the target object do not exist on the path of the moving object, the path overlap time calculation unit subtracts the predetermined value from the path overlap time. Operation support system. The operation support system according to any one of claims 2 to 7,
The route overlap time calculation unit calculates the route overlap time when the current position of the target object exists on the route of the moving object and the future position of the target object does not exist on the route of the movable object. An operation support system characterized by holding the previous value. In the operation support system according to any one of claims 2 to 8,
When the lateral movement of both the first object and the second object is detected, the course overlap time calculation unit determines that at least the future position of the target object exists on the course of the moving object. And adding the predetermined value to the course overlap time. In the operation support system according to any one of claims 2 to 8,
When the first object is recognized and the second object is not recognized, the course overlap time calculation unit determines that both the current position and the future position of the target object exist on the path of the moving object. For example, the predetermined value is added to the course overlap time. The motion support system according to any one of claims 2 to 9,
When the first object is recognized and the second object is not recognized, the course overlap time calculation unit has only one of the current position and the future position of the target object on the course of the moving object. Then, the previous value of the course overlap time is retained. The operation support system according to any one of claims 2 to 11,
When the lateral movement of the second object is detected, the lateral movement of the first object is not detected, and the lateral speed of the second object is below a predetermined speed threshold, the path overlap time calculation unit If the current position and future position of the target object exist on the path of the moving object, the predetermined value is added to the path overlap time. The operation support system according to any one of claims 2 to 12,
When the lateral movement of the second object is detected, the lateral movement of the first object is not detected, and the lateral speed of the second object is below a predetermined speed threshold, the path overlap time calculation unit The operation support system is characterized in that if only one of the current position and the future position of the target object is present on the path of the moving object, the previous value of the path overlap time is held. A position information acquisition device that acquires position information related to the position of a surrounding object existing around the moving object;
A route overlap time calculation unit that calculates a route overlap time that is a time during which the current position or future position of the surrounding object based on the position information is present on the route of the moving object, and an object recognition device comprising:
When at least one of the current position or the future position of the surrounding object is present on the path of the moving object , a predetermined value is added to the path overlap time,
The course overlap based on the position information acquisition duration during which acquisition of the position information of the peripheral object by the position information acquisition device is continued, or based on the speed of the peripheral object itself or whether the speed can be calculated. A condition for adding the predetermined value to the time is changed.

Images