Vehicle control device, control method, and control program

Abstract

To provide a vehicle control device, control method, and control program which enable calculation of an appropriate risk degree according to a surrounding situation of an own vehicle using a forward image.SOLUTION: A control device 30 includes: a first recognition unit 31 for recognizing a surrounding situation of a vehicle 1; a second recognition unit 32 for, when a shield present ahead of the vehicle 1 is recognized by the first recognition unit 31, recognizing a shielded area shielded by the shield; a determination unit 33 for determining an attribute of the shield recognized by the first recognition unit 31; a setting unit 34 for setting the shielded area shielded by the shield and a processing target area to a vehicle forward image; a calculation unit 35 for calculating a risk degree corresponding to the recognized shield on the basis of a size of an area where the shielded area and the processing target area set by the setting unit 34 overlap; and a vehicle control unit 36 for controlling the vehicle 1 on the basis of the risk degree calculated by the calculation unit 35.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a vehicle control device, a control method, and a control program for controlling a vehicle. [Background technology]

[0002] In recent years, efforts to provide access to sustainable transportation systems that take into consideration vulnerable traffic participants have been gaining momentum. As part of these efforts, research and development is being conducted on driver assistance technologies and autonomous driving technologies for automobiles and other vehicles to further improve traffic safety and convenience.

[0003] As an example of a driving assistance technology, Patent Document 1 listed below discloses a technology in which, when an obstacle is detected while a vehicle is traveling, a sudden braking speed limit point and a potential collision point are determined according to the planned route of the vehicle and position information of the obstacle, a speed limit value of the sudden braking speed limit point is calculated according to the distance between the sudden braking speed limit point and the potential collision point, and if the planned speed at the sudden braking speed limit point is equal to or less than the speed limit value, the vehicle is controlled to travel at the planned speed. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7341208 Summary of the Invention [Problem to be solved by the invention]

[0005] However, with conventional technology, it has been difficult to use an image of the front of the vehicle to calculate an appropriate risk level that is in line with the surrounding conditions of the vehicle.

[0006] The present invention provides a vehicle control device, a control method, and a control program that can calculate an appropriate risk level according to the surrounding circumstances of the vehicle using a forward image of the vehicle, thereby improving traffic safety and contributing to the development of a sustainable transportation system. [Means for solving the problem]

[0007] One aspect of the present invention is A vehicle control device that controls a vehicle, a first recognition unit that recognizes a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; a second recognition unit that recognizes an obstruction area obstructed by an obstruction present in front of the vehicle when the first recognition unit recognizes the obstruction; a discrimination unit that discriminates an attribute of the obstructing object based on a recognition result of the first recognition unit; a setting unit that sets the obstructed area and a predetermined processing target area in a forward image of the vehicle obtained based on an image capturing result of the camera, based on a recognition result of the second recognition unit and a discrimination result of the discrimination unit; a calculation unit that calculates a risk level corresponding to the obstructing object based on the size of an overlapping area between the obstructed area set by the setting unit and the processing target area; a vehicle control unit that controls the vehicle based on the risk degree calculated by the calculation unit; Equipped with The setting unit When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and The length of the processing target area in the first direction is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. A vehicle control device.

[0008] Another aspect of the present invention is The computer that controls the vehicle Recognizing a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; When an obstruction present in front of the vehicle is recognized, an obstruction area obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; based on the recognition result of the obstructed area and the determination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area are set for the image of the front of the vehicle obtained based on the image pickup result of the camera; Calculating a risk level corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Processing is performed, In the process of setting the processing target area, When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and The length of the processing target area in the first direction is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. It is a control method.

[0009] Another aspect of the present invention is The computer that controls the vehicle Recognizing a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; When an obstruction present in front of the vehicle is recognized, an obstruction area obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; based on the recognition result of the obstructed area and the determination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area are set for the image of the front of the vehicle obtained based on the image pickup result of the camera; Calculating a risk level corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Let the processing take place, In the process of setting the processing target area, When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and The length of the processing target area in the first direction is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. It is a control program. [Effects of the Invention]

[0010] According to the present invention, it is possible to provide a vehicle control device, a control method, and a control program that are capable of calculating an appropriate level of risk in accordance with the surrounding conditions of the vehicle using an image of the vehicle ahead. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a schematic configuration of a vehicle 1 equipped with a control device 30 which is an embodiment of a vehicle control device of the present invention. [Figure 2] FIG. 1 is a diagram showing an example of a situation in which an intersection obstruction SO1 that obstructs an intersection RDx, which is a road that intersects with a travel road RD on which the vehicle 1 is traveling, is present in front of the vehicle 1. [Figure 3] 3 is a diagram showing an example of a forward image FI of the vehicle 1 in the situation shown in FIG. 2. FIG. [Figure 4] FIG. 1 is a diagram showing an example of a situation in which a connecting road RDc, which is a road connected to a travel road RD on which the vehicle 1 is traveling, is located ahead of the vehicle 1 and blocks the connecting road RDc. [Figure 5] 10 is a diagram showing an example of a method for deriving an overlapping area ratio Sr between a processing target area PT and an overlapping area OA. FIG. [Figure 6] FIG. 10 is a diagram showing an example of a shielding size risk table TS used to derive a shielding size risk S. [Figure 7] FIG. 10 is a diagram showing an example of a guardrail risk table TG used to derive the guardrail risk table TG. [Figure 8] 10 is a diagram showing an example of a lane number risk table TW used to derive a lane number risk W. FIG. [Figure 9] 10 is a diagram showing an example of a pedestrian crossing risk table TC used to derive a pedestrian crossing risk C. FIG. [Figure 10] FIG. 10 is a diagram illustrating an example of a method for determining an object obstructing an intersection. [Figure 11] 4 is a flowchart showing an example of processing executed by the control device 30. [Figure 12] 11 is a flowchart (part 1) showing an example of the shielding size risk derivation process shown in FIG. [Figure 13] 11 is a flowchart (part 2) illustrating an example of the shielding size risk derivation process shown in FIG. [Figure 14] 11 is a flowchart (part 3) illustrating an example of the shielding size risk derivation process shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, an embodiment of a vehicle control device, a control method, and a control program of the present invention will be described with reference to the drawings. The drawings should be viewed in the direction of the reference symbols. Note that the following embodiment does not limit the present invention, and not all of the elements described in the following embodiment are necessarily essential to the present invention. Furthermore, two or more elements described in the following embodiment may be arbitrarily combined without departing from the spirit of the present invention. Hereinafter, identical or similar elements will be denoted by the same or similar reference symbols, and their description may be omitted or simplified.

[0013] Furthermore, in order to simplify and clarify the explanation in this specification, unless otherwise specified, the directions of front and rear (including front and rear), left and right, and up and down will be described according to the directions as seen by the driver, who is an occupant of a vehicle (i.e., vehicle 1, described later) controlled by a control device (i.e., control device 30, described later) that is one embodiment of the vehicle control device of the present invention.

[0014] [1. Vehicle] Fig. 1 is a block diagram showing a schematic configuration of a vehicle 1 equipped with a control device 30, which is one embodiment of a vehicle control device of the present invention. The vehicle 1 of this embodiment shown in Fig. 1 is an automobile equipped with a drive source (not shown) and wheels (not shown) including drive wheels driven by the power of the drive source and steerable wheels. As an example, the vehicle 1 can be a four-wheeled automobile having a pair of front wheels and a pair of rear wheels on the left and right.

[0015] The drive source of vehicle 1 may be an electric motor, an internal combustion engine such as a gasoline engine or a diesel engine, or a combination of an electric motor and an internal combustion engine. The drive source of vehicle 1 may drive a pair of left and right front wheels, a pair of left and right rear wheels, or a pair of left and right front and rear wheels, i.e., four wheels. Either one of the front wheels or the rear wheels of vehicle 1 may be a steerable wheel, or both may be steerable wheels.

[0016] The vehicle 1 is configured to include a sensor group 10, a navigation device 20, a control device 30 which is an example of a vehicle control device of the present invention, an electric power steering (EPS: Electric Power Steering) system 40, a driving force control system 50, a braking force control system 60, a communication unit 70, an operation input unit 80, and an alarm device 90.

[0017] The sensor group 10 is configured to include an external sensor 11 that acquires information about the periphery of the vehicle 1 (hereinafter also referred to as "peripheral information"), and a vehicle sensor 12 that acquires information about the vehicle 1 (hereinafter also referred to as "vehicle information"). The information acquired by each sensor included in the sensor group 10 (in other words, detected values) is output to the control device 30 and is used for controlling the vehicle 1 by the control device 30 (hereinafter also referred to as "vehicle control").

[0018] The external sensor 11 includes, for example, a camera 111, a sonar 112, and a radar 113. The camera 111 is an imaging device that captures an image of the surroundings of the vehicle 1 including the area ahead of the vehicle 1, and outputs image data of the obtained surrounding image to the control device 30. As the camera 111, for example, a digital camera using an imaging element such as a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor) can be used.

[0019] The sonar 112 emits sound waves around the vehicle 1 (for example, in front of, behind, and to the sides of the vehicle 1) and receives reflected sound from objects around the vehicle 1, thereby detecting the distance and direction of the objects. The radar 113 emits radio waves around the vehicle 1, including in front of the vehicle 1, and receives reflected waves from objects around the vehicle 1, thereby detecting the distance and direction of the objects. For example, a millimeter wave radar can be used as the radar 113.

[0020] The external sensor 11 may be configured to include a LiDAR (Light Detection and Ranging) instead of or in addition to the sonar 112 or the radar 113. In this case, the LiDAR emits laser light to the periphery of the vehicle 1 including the area ahead of the vehicle 1, and receives reflected light from an object present around the vehicle 1 to detect the distance and direction to the object.

[0021] The vehicle sensor 12 includes, for example, a wheel sensor 121, a vehicle speed sensor 122, an inertial measurement unit (IMU) 123, an occupant camera 124, an operation detection unit 125, and a steering touch sensor 126.

[0022] The wheel sensor 121 detects the rotation angle of one or more wheels of the vehicle 1. As an example, the wheel sensor 121 detects the rotation angle of each of the left rear wheel and the right rear wheel. As the wheel sensor 121, for example, an angle sensor or a displacement sensor can be adopted.

[0023] The vehicle speed sensor 122 detects the vehicle speed VP, which is the traveling speed (in other words, the moving speed of the vehicle body) of the vehicle 1. For example, the vehicle speed sensor 122 detects the vehicle speed VP based on the number of rotations of a countershaft (not shown) provided in the vehicle 1.

[0024] The inertial measurement unit 123 detects angular velocities in the pitch, roll, and yaw directions of the vehicle 1, and accelerations in the front-to-rear, left-to-right, and up-to-down directions of the vehicle 1. Note that instead of the inertial measurement unit 123, the vehicle sensor 12 may be configured to include an acceleration sensor that detects acceleration in a predetermined direction of the vehicle 1, or a gyro sensor that detects angular velocity in a predetermined direction of the vehicle 1.

[0025] The occupant camera 124 is a digital camera that captures an image of the interior of the vehicle 1 and outputs image data of the obtained interior image to the control device 30. For example, the occupant camera 124 can be a so-called "driver monitor camera" that is installed so as to be able to capture an image of the head (in other words, the face) of an occupant (hereinafter also referred to as "driver") sitting in the driver's seat of the vehicle 1 from the front. As with the camera 111, the occupant camera 124 can be a digital camera that uses an imaging element such as a CCD or CMOS.

[0026] The operation detection unit 125 detects an operation performed using the operation input unit 80 that is operable by the driver. The operation input unit 80 may include, for example, an operation button (not shown) that accepts an operation to switch a predetermined driving assistance control on (in other words, activated) and off (in other words, not activated). In this case, the operation detection unit 125 can detect an operation to turn on / off the predetermined driving assistance control. Here, an example of the predetermined driving assistance control is a collision reduction control by the vehicle control unit 36, which will be described later.

[0027] The steering touch sensor 126 detects whether the steering wheel 46 of the vehicle 1 is being properly gripped. For example, the steering touch sensor 126 is realized by a capacitance sensor or the like. In this case, the capacitance sensor is provided at a portion where the driver touches the steering wheel 46 when the steering wheel 46 is being properly gripped.

[0028] The navigation device 20 includes, for example, a GNSS (Global Navigation Satellite System) receiver 21, a touch panel 22, and a speaker 23. The navigation device 20 also has a storage unit (not shown) configured with a flash memory or the like. The storage unit of the navigation device 20 stores a map information database (DB) 24 as an example of map information.

[0029] The map information database 24 is configured to include road network information. The road network information is information that represents each road as a combination of nodes and links (also called "paths") that connect the nodes. Each node in the road network information represents a characteristic point on a road, such as an intersection, a corner, or a dead end. In the road network information, each node is set with information indicating, for example, the point corresponding to the node (for example, coordinates that can identify a point on a map, such as latitude and longitude). In addition, in the road network information, each link is set with information indicating the nodes at both ends of the link, the road corresponding to the link, the link length, the number of lanes, the direction of travel, the road type, etc.

[0030] The GNSS receiver 21 identifies the current position of the vehicle 1 (for example, the latitude and longitude of the location where the vehicle 1 is located) based on signals received from GNSS satellites. Note that the navigation device 20 may acquire, for example, detection results from the vehicle sensors 12 (for example, the wheel sensors 121 and the vehicle speed sensor 122) via the control device 30 or the like, and identify or complement the current position of the vehicle 1 by an INS (Inertial Navigation System) that uses the detection values ​​of the vehicle sensors 12.

[0031] The touch panel 22 is configured by combining a display device such as a liquid crystal display or an OLED (Organic Light Emitting Diode) with a pointing device (e.g., a touchpad). The speaker 23 is configured to be able to output audio to a passenger of the vehicle 1 (e.g., the driver).

[0032] For example, the navigation device 20 searches for a route from the current position of the vehicle 1 to a destination set by the driver using the touch panel 22 by referring to the map information database 24. Then, the navigation device 20 provides route guidance using the touch panel 22 and the speaker 23 based on the searched route. The navigation device 20 may also cause the touch panel 22 to display a predetermined information in accordance with an instruction from the control device 30. Furthermore, the navigation device 20 may output, for example, information indicating the identified current position of the vehicle 1 and predetermined information (for example, information indicating an operation received via the touch panel 22) to the control device 30.

[0033] In this embodiment, the control device 30 is configured to be able to refer to the map information database 24 (i.e., map information) of the navigation device 20. However, the present invention is not limited to this, and map information including road network information similar to that in the map information database 24 may be separately stored in the control device 30 or the like, and the control device 30 may be configured to refer to this map information.

[0034] The control device 30 is a computer that has, for example, a processor that performs various calculations, a storage unit that has a non-transitory storage medium (e.g., a flash memory) that stores various information, an input / output unit that controls input and output of data between the inside and outside of the control device 30, and the like (all not shown), and performs overall control of the vehicle 1. For example, the control device 30 is realized by one ECU (Electronic Control Unit) or by multiple ECUs working together. Note that specific examples of control by the control device 30 will be described later, so description thereof will be omitted here.

[0035] The EPS system 40 includes, for example, a steering angle sensor 41, a torque sensor 42, an EPS motor 43, a resolver 44, and an EPS ECU 45.

[0036] The steering angle sensor 41 detects the steering angle θst of the steering wheel 46 and outputs information indicating the detected steering angle θst to the EPS ECU 45. The torque sensor 42 detects the steering torque TQ, which is the torque applied to the steering wheel 46 of the vehicle 1, and outputs information indicating the detected steering torque TQ to the EPS ECU 45.

[0037] The EPS motor 43 applies a driving force or a reaction force to a steering column 47 connected to the steering wheel 46 in accordance with instructions from the EPS ECU 45, thereby assisting the driver in operating the steering wheel 46. The resolver 44 detects a rotation angle θm of the EPS motor 43, and outputs information indicating the detected rotation angle θm to the EPS ECU 45.

[0038] The EPS ECU 45 is a computer that includes, for example, a processor that performs various calculations, a storage unit that has a non-transitory storage medium that stores various information, and an input / output unit that controls input and output of data between the inside and outside of the EPS ECU 45 (all of which are not shown), and is implemented by one or more ECUs. For example, the EPS ECU 45 controls the EPS system 40 (for example, the EPS motor 43) based on the steering angle θst detected by the steering angle sensor 41, the steering torque TQ detected by the torque sensor 42, the rotation angle θm detected by the resolver 44, etc.

[0039] Furthermore, the EPS system 40 (for example, the EPS ECU 45) may output information indicating the steering angle θst detected by the steering angle sensor 41, the steering torque TQ detected by the torque sensor 42, the rotation angle θm detected by the resolver 44, etc. to the control device 30. Furthermore, the EPS system 40 (for example, the EPS ECU 45) may output information indicating the steering speed ω of the steering wheel 46 to the control device 30. In this case, the steering speed ω can be obtained, for example, by differentiating the steering angle θst with respect to time.

[0040] The driving force control system 50 includes a driving ECU 51 and is configured to be able to control the driving force of the vehicle 1. The driving ECU 51 is a computer that controls the driving force control system 50 and is realized by one or more ECUs, and includes, for example, a processor that performs various calculations, a storage unit having a non-transitory storage medium that stores various information, and an input / output unit that controls input and output of data between the inside and outside of the driving ECU 51 (all of which are not shown). For example, the driving ECU 51 controls the power output from a driving source of the vehicle 1 based on operation of an accelerator pedal 52 provided on the vehicle 1. The driving ECU 51 can also control the driving force control system 50 (for example, the driving source) according to instructions from the control device 30.

[0041] The braking force control system 60 includes a braking ECU 61 and is configured to be able to control the braking force of the vehicle 1. The braking ECU 61 is a computer that controls the braking force control system 60 and includes, for example, a processor that performs various calculations, a storage unit having a non-transitory storage medium that stores various information, and an input / output unit that controls input and output of data between the inside and outside of the braking ECU 61 (all of which are not shown), and is realized by one or more ECUs. For example, the braking ECU 61 controls the braking force of the vehicle 1 by controlling a brake device (not shown) provided in the vehicle 1 based on operation of a brake pedal 62 provided in the vehicle 1. Here, the brake device includes, for example, a brake caliper, a cylinder that transmits hydraulic pressure to the brake caliper, and an electric motor that generates hydraulic pressure in the cylinder. The braking ECU 61 controls the electric motor of the brake device so that a braking force corresponding to operation of the brake pedal 62 is generated. The braking ECU 61 can also control the braking force control system 60 (for example, the brake device) according to instructions from the control device 30.

[0042] The communication unit 70 is a communication interface that communicates with the external device 2 under the control of the control device 30. That is, the control device 30 can communicate with the external device 2 via the communication unit 70. Examples of the external device 2 include a driver's terminal device (e.g., a smartphone) and a server device managed by the manufacturer (e.g., the manufacturer) of the vehicle 1. Note that communication between the vehicle 1 and the external device 2 can be achieved using, for example, a mobile communication network such as a cellular line, Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0043] The warning device 90 is a device that issues a warning to the driver under the control of the control device 30. The warning device 90 includes, for example, an MID (Multi-Information Display) 91 and a buzzer 92.

[0044] The MID 91 is configured with a display device such as a liquid crystal display or an OLED, and is provided in a position visible to the driver (for example, in the meter panel of the vehicle 1). For example, the MID 91 displays a predetermined warning image in accordance with an instruction from the control device 30. The MID 91 may be shared with the above-mentioned touch panel 22. That is, the term "MID 91" in the following description may be read as "touch panel 22."

[0045] The buzzer 92 is configured to be able to output a predetermined alarm sound. For example, the buzzer 92 outputs the predetermined alarm sound in accordance with an instruction from the control device 30. The buzzer 92 may be shared with the above-mentioned speaker 23. That is, the "buzzer 92" in the following description may be read as the "speaker 23."

[0046] [2. Control device] Next, a more detailed description will be given of the control device 30. The control device 30 includes, for example, a first recognition unit 31, a second recognition unit 32, a discrimination unit 33, a setting unit 34, a calculation unit 35, and a vehicle control unit 36 ​​as functional units realized by, for example, a processor executing a program stored in a storage unit of the control device 30.

[0047] The first recognition unit 31 recognizes the surrounding situation of the vehicle 1 based on surrounding information obtained by the external sensor 11 including the camera 111 that captures at least an image ahead of the vehicle 1. For example, the first recognition unit 31 performs sensor fusion processing on the detection results from some or all of the camera 111, sonar 112, and radar 113 included in the external sensor 11, and recognizes the surrounding situation of the vehicle 1 based on the processing results.

[0048] The first recognition unit 31 recognizes the position, type, speed, acceleration, etc. of objects present around the vehicle 1 as the surrounding conditions of the vehicle 1. At this time, the first recognition unit 31 recognizes the position of the object as a position on an absolute coordinate system with a representative point of the vehicle 1 (for example, the center of gravity or the center of the drive shaft) as the origin. This makes it possible to recognize the relative position between the vehicle 1 and the objects present around it. Furthermore, on the above absolute coordinate system, the position of the object may be represented using a representative point such as the center of gravity or a corner of the object, or may be represented as an area.

[0049] Examples of objects that can be recognized by the first recognition unit 31 include traffic participants such as other vehicles and pedestrians, road boundaries such as lane markings, curbs, and median strips, road structures such as guardrails and road shoulders, and features such as buildings, fences, and trees. The first recognition unit 31 may also recognize other road phenomena such as crosswalks, stop lines, traffic lights, road signs, forks, merging points, interchanges, and toll booths.

[0050] The first recognition unit 31 can recognize, for example, the shape of the road (hereinafter also referred to as "travel path RD") on which the vehicle 1 travels. Furthermore, the first recognition unit 31 can recognize road phenomena such as traffic participants (e.g., other vehicles) present around the vehicle 1, features (e.g., buildings) present around the vehicle 1, structures (e.g., guardrails) provided at the boundary between the travel path RD and its outside to prevent entry onto the travel path RD, and crosswalks and traffic lights present around the vehicle 1.

[0051] When the first recognition unit 31 recognizes an obstruction (hereinafter also referred to as "obstruction SO") present ahead of the vehicle 1, the second recognition unit 32 recognizes an obstructed area (hereinafter also referred to as "obstructed area SA") obstructed by the obstruction SO. In this embodiment, for example, features such as buildings and other vehicles present around the vehicle 1 are recognized as the obstruction SO. Then, the second recognition unit 32 recognizes an area that is shaded by the object recognized as the obstruction SO as the obstructed area SA caused by the obstruction SO, as viewed from the vehicle 1.

[0052] The discrimination unit 33 discriminates the attributes of the obstructing object SO based on the recognition result of the first recognition unit 31. For example, the discrimination unit 33 discriminates the attributes of the obstructing object SO based on the position where the obstructing object SO is present and the type of object that serves as the obstructing object SO. In this embodiment, the discrimination unit 33 discriminates, as the attributes of the obstructing object SO, for example, whether the obstructing object SO is an intersection obstructing object (hereinafter also referred to as "intersection obstructing object SO1"), a connecting road obstructing object (hereinafter also referred to as "connecting road obstructing object SO2"), or a non-intersection obstructing object (hereinafter also referred to as "non-intersection obstructing object SO3").

[0053] Here, the intersection obstruction SO1 is an obstruction that obstructs at least a part of an intersection (hereinafter also referred to as "intersection RDx"), which is a road that intersects with the travel path RD ahead of the vehicle 1 (see, for example, FIG. 2). For example, the discrimination unit 33 may discriminate, as the intersection obstruction SO1, an object such as a building or a fence that is present on the near side of the intersection RDx as seen from the vehicle 1. An example of a method for discriminating the intersection obstruction SO1 will be described later.

[0054] The connecting road obstruction SO2 is an obstruction that obstructs at least a part of a connecting road (hereinafter also referred to as "connecting road RDc"), which is a road that connects to the travel road RD ahead of the vehicle 1 (see, for example, FIG. 4). For example, the discrimination unit 33 can discriminate, as the connecting road obstruction SO2, an object on the ground such as a building or a fence that is present on the near side of the connecting road RDc as seen from the vehicle 1.

[0055] In addition, the non-intersection road obstruction SO3 is an obstruction different from the intersection road obstruction SO1 and the connecting road obstruction SO2. For example, the discrimination unit 33 can discriminate another vehicle parked in front of the vehicle 1 as the non-intersection road obstruction SO3.

[0056] Furthermore, the discrimination unit 33 may be configured to determine, as an attribute of the obstructing object SO, whether the obstructing object SO is a vehicle. Furthermore, the discrimination unit 33 may be configured to determine, as an attribute of the obstructing object SO, whether the obstructing object SO is an oncoming vehicle. Here, an oncoming vehicle is another vehicle facing in the opposite direction from the vehicle 1 (in other words, another vehicle traveling in the opposite direction from the vehicle 1). By having the discrimination unit 33 also determine whether the obstructing object SO is a vehicle (for example, an oncoming vehicle), it becomes possible to set an appropriate processing target region PT (described later) even when the obstructing object SO is a vehicle.

[0057] If the control device 30 is configured to be able to refer to map information such as the map information database 24, the determination unit 33 may determine the attributes of the obstructing object SO based on the recognition result of the first recognition unit and the map information. In this case, the determination unit 33, for example, first identifies the location (e.g., latitude and longitude) where the obstructing object SO is located based on the relative position between the obstructing object SO and the vehicle 1 recognized by the first recognition unit 31 and the current location of the vehicle 1 identified by the navigation device 20 (e.g., the GNSS receiver 21). Then, the determination unit 33 may determine the attributes of the obstructing object SO based on the location where the obstructing object SO is located, by referring to road network information included in the map information.

[0058] In this way, by determining the attributes of the obstructing object SO using map information in addition to the recognition results of the first recognition unit 31, it is possible to determine the attributes of the obstructing object SO more accurately than when the attributes of the obstructing object SO are determined based solely on the recognition results of the first recognition unit 31.

[0059] Based on the recognition result of the second recognition unit 32 and the discrimination result of the discrimination unit 33, the setting unit 34 sets a shielded area SA and a predetermined processing target area PT in a forward image of the vehicle 1 (hereinafter also referred to as a "forward image FI") obtained based on the imaging result of the camera 111. Here, the forward image FI is, for example, an image obtained as an imaging result of an imaging device that images the area in front of the vehicle 1, but is not limited to this and may be an image obtained by combining imaging results of imaging devices that image the area in front of the vehicle 1 and the left and right sides thereof. Note that an example of a specific method for setting the shielded area SA and the processing target area PT will be described later, and therefore a description thereof will be omitted here.

[0060] The calculation unit 35 calculates the risk level (hereinafter also referred to as the "risk level VR") corresponding to the obstruction SO that creates the obstruction area SA set by the setting unit 34, based on the size of the overlapping area (hereinafter also referred to as the "overlapping area OA") between the obstruction area SA and the processing target area PT.

[0061] Here, the risk degree VR is an evaluation value that represents the magnitude of the risk of collision between the vehicle 1 and a moving object (e.g., another vehicle or a pedestrian) that may enter the driving path RD from behind the obstruction SO (in other words, from the area SA blocked by the obstruction SO). In other words, the greater the expected risk, the greater the risk degree VR. Note that an example of a specific method for calculating the risk degree VR will be described later, and therefore will not be described here.

[0062] The vehicle control unit 36 ​​controls the vehicle 1 (e.g., acceleration / deceleration of the vehicle 1) based on the risk degree VR calculated by the calculation unit 35. In this embodiment, the vehicle control unit 36 ​​executes collision reduction control. Specifically, for example, if the calculated risk degree VR is equal to or greater than a predetermined first threshold and less than a predetermined second threshold (where the second threshold is greater than the first threshold), the vehicle control unit 36 ​​issues a warning to alert the driver via the warning device 90 before the vehicle 1 reaches the vicinity of the obstruction SO corresponding to the risk degree VR. Furthermore, if the calculated risk degree VR is equal to or greater than the second threshold, in addition to issuing the warning, the vehicle control unit 36 ​​brakes the vehicle 1 via the braking force control system 60 so that the vehicle 1 stops at a position in front of the obstruction SO or obstructed area SA corresponding to the risk degree VR. The first threshold and the second threshold are determined in advance by the manufacturer of the vehicle 1. The above-mentioned warning can be realized, for example, by displaying a predetermined warning image on the MID 91 or by outputting a predetermined warning sound from the buzzer 92.

[0063] (2-1. Example of how to set the masked area and the processing target area) Next, an example of a specific method for setting the masked area SA and the processing target area PT in the forward image FI will be described.

[0064] (In case of intersection obstruction) In the example shown in Fig. 2, an obstruction SO (e.g., a building) is present just before the left side of the intersection RDx that intersects with the travel path RD (in other words, on the front left side of the vehicle 1). Also, in the example shown in Fig. 2, the vehicle 1 is traveling on the travel path RD toward the intersection of the travel path RD and the intersection RDx.

[0065] Fig. 3 shows an example of a forward image FI in the situation shown in Fig. 2. The left-right direction (direction indicated by LR in Fig. 3) in the forward image FI shown in Fig. 3 corresponds to the left-right direction of the vehicle 1 (in other words, the width direction of the vehicle 1), and is an example of a first direction in the present invention. Moreover, the up-down direction (direction indicated by UD in Fig. 3) in the forward image FI shown in Fig. 3 corresponds to the up-down direction of the vehicle 1, and is an example of a second direction in the present invention.

[0066] 2, the first recognition unit 31 recognizes an obstruction SO located in front of the left side of the intersection RDx, and the second recognition unit 32 recognizes a blocked area SA caused by the obstruction SO. Furthermore, the discrimination unit 33 discriminates the obstruction SO as an intersection obstruction SO1 because the obstruction SO obstructs the left side of the intersection RDx.

[0067] Then, the setting unit 34 sets, in the forward image FI, a blocked area SA caused by the obstruction SO determined to be the intersection obstruction SO1. For example, as shown by the dotted hatching in Fig. 3, the setting unit 34 sets, as the blocked area SA, an area on the forward image FI corresponding to the obstruction SO determined to be the intersection obstruction SO1. The setting unit 34 also sets, in the forward image FI, a processing target area PT used for calculating the risk level VR corresponding to the obstruction SO determined to be the intersection obstruction SO1.

[0068] For example, as shown in FIG. 2, at an intersection RDx shielded by an intersection shield SO1, there may be a virtual moving body VM, such as another vehicle, that may enter the travel path RD from the intersection RDx.

[0069] Therefore, in such a case, when setting the processing target area PT, the setting unit 34 first derives a virtual collision point CP, which is a virtual collision point between the vehicle 1 and the virtual moving body VM, on the assumption that the virtual moving body VM enters the traveling path RD from the intersection RDx when the vehicle 1 reaches the intersection of the traveling path RD and the intersection RDx. The setting unit 34 may derive the virtual collision point CP based on the shape or traveling line of the vehicle 1, or may simply derive an arbitrary point within the intersection of the traveling path RD and the intersection RDx (for example, the entrance or center of the intersection on the traveling path RD side) as the virtual collision point CP.

[0070] 3, the setting unit 34 sets, as the processing target region PT, a region in the forward image FI that extends toward the intersection obstruction SO1 in the left-right direction (i.e., the direction corresponding to the width direction of the vehicle 1) with the virtual collision point CP as a reference and has a predetermined height h in the up-down direction (i.e., the direction corresponding to the up-down direction of the vehicle 1). In the following description, the height h (i.e., the length in the up-down direction) of the processing target region PT in the forward image FI is also referred to as the "vertical distance h of the processing target region PT." Furthermore, the left-right length of the processing target region PT in the forward image FI is also referred to as the "horizontal distance d1 of the processing target region PT."

[0071] The vertical distance h of the processing target region PT can be determined arbitrarily by the manufacturer of the vehicle 1, but can be set to, for example, a height corresponding to a distance of 2 m above the ground at the virtual collision point CP. This takes into consideration the fact that an object present within a distance of 2 m above the ground at the virtual collision point CP may collide with the vehicle 1, but an object present above that distance is unlikely to collide with the vehicle 1.

[0072] In this way, by determining the vertical distance h of the processing target area PT taking into account the range in which an object that may collide with the vehicle 1 may exist, it is possible to set the processing target area PT with an appropriate vertical distance h without making the processing target area PT excessively large in the vertical direction.

[0073] Furthermore, if the obstruction SO is determined to be an intersection obstruction SO1, the setting unit 34 sets the lateral distance d1 of the processing target area PT to a length corresponding to a distance according to the minimum braking time Tmin and a predetermined first speed.

[0074] Here, the minimum braking time Tmin is the time required for the vehicle 1 to stop when it is decelerated from the current vehicle speed VP (i.e., traveling speed) at a predetermined deceleration x. The deceleration x can be the deceleration that occurs when the vehicle 1 applies the so-called "sudden brakes," and can be set to 0.6 G (here, G is the acceleration due to gravity) as an example. In the following description, the braking distance of the vehicle 1 when it is decelerated from the current vehicle speed VP at the deceleration x is also referred to as the "minimum braking distance Dmin."

[0075] As shown in Figure 2, when a virtual moving body VM exists at an intersection RDx shielded by an intersection shield SO1, this virtual moving body VM may be moving at a speed of about 30 km / h on the intersection RDx toward the intersection of the travel path RD and the intersection RDx. Also, as shown in Figure 2, in the case of the intersection RDx, the intersection of the travel path RD and the intersection RDx forms a cross shape, so a virtual moving body VM that may exist on the intersection RDx can go straight through the intersection of the travel path RD and the intersection RDx. Therefore, in this case, there is a risk that the virtual moving body VM moving at a speed of about 30 km / h may enter the travel path RD from the intersection RDx shielded by the intersection shield SO1.

[0076] In order for the vehicle 1 to avoid a collision with such a virtual moving object VM, it is desirable that the driver be able to see the range from the virtual collision point CP to a point in the direction in which the virtual moving object VM could be located, the range being a virtual movement distance Dv, which is the distance the virtual moving object VM can move in the minimum braking time Tmin, before the vehicle 1 reaches the collision avoidance limit point LP, which is a point on the driving path RD that is the minimum braking distance Dmin before the virtual collision point CP. In other words, the range from the virtual collision point CP to a point in the direction in which the virtual moving object VM could be located, the virtual movement distance Dv, can be said to be the "range that the driver should originally be able to see."

[0077] Therefore, when the obstacle SO is determined to be an intersection obstacle SO1, the setting unit 34 sets the first speed to, for example, 30 [km / h] and the lateral distance d1 of the processing target region PT to a length corresponding to the distance expressed by the product of the minimum braking time Tmin and 30 [km / h]. That is, by setting the first speed to 30 [km / h], which is considered to be the moving speed of a virtual moving object VM that may be present at the intersection RDx, when an intersection obstacle SO1 that blocks the intersection RDx is present ahead of the vehicle 1, the processing target region PT can be set to have a lateral distance d1 corresponding to the virtual moving distance Dv of the virtual moving object VM that may be present at the intersection RDx. In other words, when an intersection obstacle SO1 that blocks the intersection RDx is present ahead of the vehicle 1, the processing target region PT can be set to have a lateral distance d1 corresponding to the "range that the driver should be able to see." Therefore, it is possible to set the processing target region PT to have an appropriate lateral distance d1 that is in line with the surrounding conditions of the vehicle 1.

[0078] In the example described here, the first speed is set to 30 km / h, but is not limited to this. The first speed can be set arbitrarily by, for example, the manufacturer of the vehicle 1, and may be smaller or larger than 30 km / h. Furthermore, if the control device 30 can refer to map information such as the map information database 24, the setting unit 34 may refer to this map information and use, as the first speed, the legal speed limit for the intersection RDx that is blocked by the intersection obstruction SO1.

[0079] As described above, when the obstacle SO is determined to be the intersection obstacle SO1, the setting unit 34 can set, as a processing target area PT, an area in the forward image FI extending toward the intersection obstacle SO1 in the left-right direction based on the virtual collision point CP between the vehicle 1 and a virtual moving object VM that may be approaching the road RD from the intersection RDx on the intersection obstacle SO1 side at a first speed (e.g., 30 km / h). The lateral distance d1, which is the length of the processing target area in this case, can be set to a length corresponding to the minimum braking time Tmin, which is the braking time when the vehicle 1 is decelerated at a predetermined deceleration x, and the distance according to the first speed. This makes it possible to set a processing target area PT having an appropriate lateral distance d1 in accordance with the surrounding conditions of the vehicle 1.

[0080] Therefore, the control device 30 can calculate an appropriate risk level VR that is suited to the surrounding conditions of the vehicle 1, assuming a configuration in which the risk level VR corresponding to an obstacle SO present ahead of the vehicle 1 is calculated using the forward image FI. Specifically, when an intersection obstacle SO1 is present ahead of the vehicle 1, the control device 30 can calculate the risk level VR that is suited to the surrounding conditions of the vehicle 1, taking into account a virtual moving object VM that may enter the road RD from the intersection RDx blocked by the intersection obstacle SO1 at a first speed, as the risk level VR corresponding to the intersection obstacle SO1. This makes it possible to calculate an appropriate risk level VR that is suited to the surrounding conditions of the vehicle 1 with a simple configuration, and to appropriately control the vehicle 1 based on the risk level VR. This can ultimately contribute to the development of a sustainable transportation system.

[0081] Furthermore, the setting unit 34 can set a rectangular processing target region PT that extends in the left-right direction and has a predetermined height h in the up-down direction, which simplifies the setting of the processing target region PT and reduces the processing load on the control device 30 when setting the processing target region PT.

[0082] (In case of connection path obstruction) In the example shown in Fig. 4, a connecting road RDc exists instead of the intersection RDx in the example shown in Fig. 2. Here, the connecting road RDc is a road that connects to the travel road RD ahead of the vehicle 1, and in the example shown in Fig. 4, it is connected to the left side of the travel road RD ahead of the vehicle 1. Also, in the example shown in Fig. 4, there is an obstruction SO (e.g., a building) in front of the connecting road RDc, and the vehicle 1 is traveling on the travel road RD toward the connection point between the travel road RD and the connecting road RDc.

[0083] 4, as in the situation shown in Fig. 2, the first recognition unit 31 recognizes an obstruction SO that exists in front of the connecting road RDc, and the second recognition unit 32 recognizes a blocked area SA caused by this obstruction SO. Furthermore, because this obstruction SO obstructs the connecting road RDc, the discrimination unit 33 discriminates this obstruction SO as a connecting road obstruction SO2. Then, the setting unit 34 sets, in the forward image FI, the blocked area SA caused by the obstruction SO that has been discriminated to be a connecting road obstruction SO2 and a processing target area PT.

[0084] For example, as shown in Fig. 4, on the connecting road RDc shielded by the connecting road obstruction SO2, there may be a virtual moving object VM that may enter the traveling road RD from the connecting road RDc. In such a case, when setting the processing target area PT, the setting unit 34 first derives a virtual collision point CP between the vehicle 1 and the virtual moving object VM on the assumption that the virtual moving object VM enters the traveling road RD from the connecting road RDc when the vehicle 1 reaches the junction between the traveling road RD and the connecting road RDc. Then, the setting unit 34 sets, as the processing target area PT, an area in the forward image FI that extends from the virtual collision point CP in the left-right direction toward the connecting road obstruction SO2 and has a predetermined height h in the up-down direction.

[0085] The vertical distance h of the processing target area PT set when the obstruction SO is determined to be a connecting road obstruction SO2 can also be set to a height corresponding to a distance of 2 m above the ground at the virtual collision point CP, similar to the vertical distance h of the processing target area PT set when the obstruction SO is determined to be an intersection road obstruction SO1.

[0086] On the other hand, it is preferable that the horizontal distance d1 of the processing target area PT set when the obstruction SO is determined to be a connecting road obstruction SO2 is shorter than the horizontal distance d1 of the processing target area PT set when the obstruction SO is determined to be an intersection obstruction SO1.

[0087] That is, as described above, in the case of the intersection road RDx, the intersection of the travel road RD and the intersection road RDx is cross-shaped, so a virtual moving object VM that may be present on the intersection road RDx can go straight through the intersection of the travel road RD and the intersection road RDx. In contrast, as shown in FIG. 4, in the case of the connecting road RDc, the connection point between the travel road RD and the connecting road RDc is T-shaped, so a virtual moving object VM that may be present on the connecting road RDc cannot go straight through the connection point between the travel road RD and the connecting road RDc. Therefore, it is considered that a virtual moving object VM that attempts to enter the travel road RD from the connecting road RDc will decelerate before entering the travel road RD. In other words, when the obstacle SO is the connecting road obstacle SO2, the virtual moving distance Dv that the virtual moving object VM can travel in the minimum braking time Tmin (in other words, the driver's "normal viewing range") may be shorter than when the obstacle SO is the intersection road obstacle SO1.

[0088] Therefore, when the obstacle SO is determined to be a connecting path obstacle SO2, the setting unit 34 sets the lateral distance d1 of the processing target area PT to a length corresponding to a distance determined by the minimum braking time Tmin and a third speed lower than the first speed. More specifically, in this case, the setting unit 34 sets the lateral distance d1 of the processing target area PT to a length corresponding to a distance represented by the product of the minimum braking time Tmin and the third speed. As a result, when a connecting path obstacle SO2 blocking the connecting path RDc is present ahead of the vehicle 1, the processing target area PT can be set to have a lateral distance d1 corresponding to the virtual movement distance Dv of a virtual moving object VM that may be present on the connecting path RDc. In other words, when a connecting path obstacle SO2 blocking the connecting path RDc is present ahead of the vehicle 1, the processing target area PT can be set to have a lateral distance d1 corresponding to the driver's "intended viewing range." Therefore, it is possible to set the processing target area PT to have an appropriate lateral distance d1 that is in line with the surrounding conditions of the vehicle 1.

[0089] The third speed can be set arbitrarily by the manufacturer of the vehicle 1, and can be set to, for example, 15 km / h. The third speed is not limited to 15 km / h, and may be smaller or larger than 15 km / h. If the control device 30 can refer to map information such as the map information database 24, the setting unit 34 may refer to this map information and determine the third speed while taking into account the legal speed limit of the connecting road RDc that is blocked by the connecting road obstruction SO2.

[0090] As described above, when the obstacle SO is determined to be the connecting road obstacle SO2, the setting unit 34 can set, as a processing target area PT, an area in the forward image FI extending toward the connecting road obstacle SO2 in the left-right direction based on the virtual collision point CP between the vehicle 1 and a virtual moving object VM that may be entering the traveling road RD from the connecting road RDc at a third speed (e.g., 15 km / h). The lateral distance d1, which is the length of the processing target area in the left-right direction, can be set to a length corresponding to the minimum braking time Tmin, which is the braking time when the vehicle 1 is decelerated at a predetermined deceleration x, and the distance corresponding to the third speed. This makes it possible to set a processing target area PT having an appropriate lateral distance d1 that is suited to the surrounding conditions of the vehicle 1.

[0091] Therefore, the control device 30 can calculate an appropriate risk degree VR that is suited to the surrounding conditions of the vehicle 1, assuming a configuration in which the risk degree VR corresponding to an obstacle SO present ahead of the vehicle 1 is calculated using the forward image FI. Specifically, when a connecting path obstacle SO2 is present ahead of the vehicle 1, the control device 30 can calculate the risk degree VR that is suited to the surrounding conditions of the vehicle 1, taking into account a virtual moving object VM that may enter the traveling path RD from the connecting path RDc that is blocked by the connecting path obstacle SO2 at a third speed, as the risk degree VR corresponding to the connecting path obstacle SO2. This makes it possible to calculate an appropriate risk degree VR that is suited to the surrounding conditions of the vehicle 1 with a simple configuration, and to appropriately control the vehicle 1 based on the risk degree VR.

[0092] (In case of non-crossing road cover) Even when the discrimination unit 33 discriminates that the obstruction SO is a non-intersection obstruction SO3, the setting unit 34 sets the obstruction area SA and the processing target area PT caused by the obstruction SO discriminated as a non-intersection obstruction SO3 in the forward image FI.

[0093] That is, in such a case, the setting unit 34 derives a virtual collision point CP between the vehicle 1 and the virtual moving body VM on the assumption that the virtual moving body VM has entered the travel road RD from behind the non-intersection road obstruction SO3 (i.e., the obstructed area SA by the non-intersection road obstruction SO3).The setting unit 34 then sets an area extending from the virtual collision point CP in the left-right direction toward the non-intersection road obstruction SO3 and having a predetermined height h in the up-down direction as the processing target area PT.

[0094] The vertical distance h of the processing target area PT that is set when the obstruction SO is determined to be a non-intersection obstruction SO3 can also be set to a height corresponding to a distance of 2 m above the ground at the virtual collision point CP, similar to the vertical distance h of the processing target area PT that is set when the obstruction SO is determined to be an intersection obstruction SO1 or a connecting road obstruction SO2.

[0095] On the other hand, it is preferable that the horizontal distance d1 of the processing target area PT set when the obstruction SO is determined to be a non-intersection obstruction SO3 is shorter than the horizontal distance d1 of the processing target area PT set when the obstruction SO is determined to be an intersection obstruction SO1 or a connecting road obstruction SO2.

[0096] That is, as described above, the discrimination unit 33 can discriminate another vehicle or the like stopped ahead of the vehicle 1 as a non-intersection road obstruction SO3. Therefore, a moving object that may enter the travel road RD from behind the obstruction SO that has been discriminated as a non-intersection road obstruction SO3 is highly likely to be, for example, a slow-moving pedestrian or bicycle.

[0097] Therefore, when the obstacle SO is determined to be a non-intersection road obstacle SO3, the setting unit 34 sets the lateral distance d1 of the processing target area PT to a length corresponding to the distance determined by the minimum braking time Tmin and a second speed that is lower than the first and third speeds. More specifically, in this case, the setting unit 34 sets the lateral distance d1 of the processing target area PT to a length corresponding to the distance represented by the product of the minimum braking time Tmin and the second speed. This allows the processing target area PT to be set with a lateral distance d1 corresponding to the virtual movement distance Dv of a virtual moving object VM that may be present in the area SA blocked by the non-intersection road obstacle SO3, when the non-intersection road obstacle SO3 is present ahead of the vehicle 1. In other words, when the non-intersection road obstacle SO3 is present ahead of the vehicle 1, the processing target area PT can be set with a lateral distance d1 corresponding to the driver's "intended viewing range." This makes it possible to set the processing target area PT with an appropriate lateral distance d1 that is appropriate for the surrounding conditions of the vehicle 1.

[0098] The second speed can be set arbitrarily by the manufacturer of the vehicle 1, and can be set to, for example, 5 km / h. The second speed is not limited to 5 km / h, and may be smaller or larger than 5 km / h.

[0099] As described above, when the obstacle SO is determined to be a non-intersection road obstacle SO3, the setting unit 34 can set, as a processing target area PT, an area in the forward image FI extending in the left-right direction toward the non-intersection road obstacle SO3, based on the virtual collision point CP between the vehicle 1 and a virtual moving object VM that may enter the road RD from the blocked area SA by the non-intersection road obstacle SO3 at a second speed (for example, 5 km / h). In this case, the lateral distance d1, which is the length in the left-right direction of the processing target area PT, can be set to a length corresponding to the minimum braking time Tmin, which is the braking time when the vehicle 1 is decelerated at a predetermined deceleration x, and the distance according to the second speed.

[0100] Therefore, the control device 30 can calculate an appropriate risk degree VR that is suited to the surrounding conditions of the vehicle 1, assuming a configuration in which the risk degree VR corresponding to the obstacle SO present ahead of the vehicle 1 is calculated using the forward image FI. Specifically, when a non-intersection road obstacle SO3 is present ahead of the vehicle 1, the control device 30 can calculate the risk degree VR that is suited to the surrounding conditions of the vehicle 1, taking into account a virtual moving object VM that may enter the road RD from the blocked area SA blocked by the non-intersection road obstacle SO3 at a second speed, as the risk degree VR corresponding to the non-intersection road obstacle SO3. This makes it possible to calculate an appropriate risk degree VR that is suited to the surrounding conditions of the vehicle 1 with a simple configuration, and to appropriately control the vehicle 1 based on the risk degree VR.

[0101] (2-2. Example of how to calculate the degree of risk) Next, an example of a specific method for calculating the risk level VR will be described. The calculation unit 35 derives various parameters, such as the obstruction size risk S, the guardrail risk G, the number of lanes risk W, and the pedestrian crossing risk C, and calculates the risk level VR based on these derived parameters.

[0102] (Shielding size risk S) First, an example of a method for deriving the occlusion size risk S will be described with reference to Figures 3, 5, and 6. When deriving the occlusion size risk S, the calculation unit 35 first derives an overlapping area ratio Sr, which is the ratio of the overlapping area OA to the processing target area PT, based on the size of the processing target area PT and the size of the overlapping area OA. The processing target area PT can be said to be the "area that should be seen" by the driver. Furthermore, the overlapping area OA, where the processing target area PT and the occlusion area SA overlap, can be said to be the "area that is not visible" to the driver. Therefore, the overlapping area ratio Sr can also be said to be the "ratio of the size of the invisible area to the size of the area that should be visible."

[0103] 3 and 5, the size of the processing target region PT can be calculated by multiplying the vertical distance h of the processing target region PT by the horizontal distance d1 of the processing target region PT. In other words, the size of the processing target region PT can also be said to be the area of ​​the processing target region PT in the forward image FI.

[0104] 3 and 5, when the overlapping area OA is rectangular, the size of the overlapping area OA can be calculated by multiplying the height of the overlapping area OA in the forward image FI (in other words, the vertical length; hereinafter also referred to as the "vertical distance of the overlapping area OA") by the horizontal length of the overlapping area OA in the forward image FI (hereinafter also referred to as the "horizontal distance d11 of the overlapping area OA"). In other words, the size of the overlapping area OA can also be said to be the area of ​​the overlapping area OA in the forward image FI.

[0105] Therefore, as shown in Figures 3 and 5, for example, if the vertical distance between the processing target area PT and the overlapping area OA is h, the horizontal distance between the processing target area PT is d1, and the horizontal distance between the overlapping area OA is d11, the overlapping area ratio Sr can be calculated, for example, using the following equation (1).

[0106] Overlap area ratio Sr = (h × d11) / (h × d1) (1)

[0107] In the example described here, the overlapping area ratio Sr is calculated from the size of the processing target area PT and the size of the overlapping area OA, but this is not limiting. For example, the calculation unit 35 may calculate the overlapping area ratio Sr as the ratio between the size of the overlapping area OA and the size of the non-overlapping area NA, which is the area remaining after excluding the overlapping area OA from the processing target area PT.

[0108] After calculating the overlapping area ratio Sr, the calculation unit 35 derives the occlusion size risk S based on the calculated overlapping area ratio Sr, for example, by referring to an occlusion size risk table TS shown in Fig. 6. Here, the occlusion size risk table TS is a table (i.e., information) that defines the occlusion size risk S according to the range of the overlapping area ratio Sr, and is stored in advance in a storage unit of the control device 30, etc.

[0109] In the occlusion size risk table TS shown in Fig. 6, the occlusion size risk S of "S1" corresponds to the overlapping area ratio range of "Sr1 or more," the occlusion size risk S of "S2" corresponds to the overlapping area ratio range of "Sr2 or more and less than Sr1," the occlusion size risk S of "S3" corresponds to the overlapping area ratio range of "Sr3 or more and less than Sr2," and the occlusion size risk S of "S4" corresponds to the overlapping area ratio range of "less than Sr3." Here, S1, S2, S3, and S4 are each predetermined values, and as shown in Fig. 6, there is a relationship of S1>S2>S3>S4.

[0110] For example, when the calculation unit 35 determines that the calculated overlapping area ratio Sr is a value of "Sr1 or more," it refers to the occlusion size risk table TS and derives "S1" corresponding to the overlapping area ratio range of "Sr1 or more" as the occlusion size risk S. Furthermore, when the calculation unit 35 determines that the calculated overlapping area ratio Sr is a value of "Sr2 or more and less than Sr1," it refers to the occlusion size risk table TS and derives "S2" corresponding to the overlapping area ratio range of "Sr2 or more and less than Sr1" as the occlusion size risk S.

[0111] In addition, when an intersection obstruction SO1 as shown in Figure 2 exists, the horizontal distance d1 of the processing target area PT is larger than when a connecting road obstruction SO2 as shown in Figure 4 exists or when a non-intersection road obstruction SO3 exists, and therefore the derived obstruction size risk S is also likely to be larger.

[0112] (Guardrail Risk G) Next, an example of a method for deriving the guard rail risk G will be described. The calculation unit 35 derives the guard rail risk G based on the recognition result of the first recognition unit 31, for example, by referring to the guard rail risk table TG shown in FIG. 7. Here, the guard rail risk table TG is a table (i.e., information) that defines the guard rail risk G according to the installation status of the guard rail, and is stored in advance in the storage unit of the control device 30 or the like.

[0113] In the guard rail risk table TG shown in FIG. 7, for the installation status of the guard rail of "none on both sides", the guard rail risk G of "G1" is corresponding, for the installation status of the guard rail of "only on the left side", the guard rail risk G of "G2" is corresponding, for the installation status of the guard rail of "only on the right side", the guard rail risk G of "G3" is corresponding, and for the installation status of the guard rail of "on both sides", the guard rail risk G of "G4" is corresponding, respectively. Here, G1, G2, G3, and G4 are predetermined values defined in advance, and as shown in FIG. 7, they have the relationship of G1 < G2 < G3 < G4.

[0114] For example, when the calculation unit 35 determines based on the recognition result of the first recognition unit 31 that there is no guard rail on both sides of the traveling road RD, it refers to the guard rail risk table TG and derives "G1" corresponding to the installation status of the guard rail of "none on both sides" as the guard rail risk G. Further, for example, when the calculation unit 35 determines based on the recognition result of the first recognition unit 31 that there is only a guard rail on the left side of the traveling road RD, it refers to the guard rail risk table TG and derives "G2" corresponding to the installation status of the guard rail of "only on the left side" as the guard rail risk G.

[0115] According to the guardrail risk table TG shown in Fig. 7, when the guardrail installation situation is "none on both sides," the maximum value can be derived as the guardrail risk G. Also, when the guardrail installation situation is "present on both sides," the minimum value can be derived as the guardrail risk G. This is because it is assumed that on roads where structures such as guardrails that prevent people from entering the road are installed, there are fewer pedestrians and other moving objects that may unexpectedly enter the road compared to roads where such structures are not installed.

[0116] Furthermore, according to the guardrail risk table TG shown in Fig. 7, when the guardrail installation situation is "on the right side only," a smaller value can be derived as the guardrail risk G compared to when the guardrail installation situation is "on the left side only." This is because, assuming left-hand traffic, when the guardrail is installed farther from vehicle 1 (i.e., on the right), the distance between vehicle 1 and a moving object such as a pedestrian climbing over the guardrail and entering the road is greater than when the guardrail is installed closer to vehicle 1 (i.e., on the left). This means that it is assumed that there is a lower possibility of an immediate collision with vehicle 1.

[0117] (Number of lanes risk W) Next, an example of a method for deriving the lane number risk W will be described. The calculation unit 35 derives the lane number risk W based on the recognition result of the first recognition unit 31, for example, by referring to the lane number risk table TW shown in Fig. 8. Here, the lane number risk table TW is a table (i.e., information) that defines the lane number risk W according to the number of lanes (in other words, road width), and is stored in advance in a memory unit of the control device 30, etc.

[0118] In the lane number risk table TW shown in Fig. 8, a lane number of "1 lane" is associated with a lane number risk W of "W1," a lane number of "2 lanes" is associated with a lane number risk W of "W2," a lane number of "3 lanes" is associated with a lane number risk W of "W3," a lane number of "4 lanes" is associated with a lane number risk W of "W4," and a lane number of "5 or more lanes" is associated with a lane number risk W of "W5." Here, W1, W2, W3, W4, and W5 are each predetermined values, and as shown in Fig. 8, there is a relationship of W1>W2>W3>W4>W5.

[0119] For example, when the calculation unit 35 determines that the number of lanes on the travel road RD is "1 lane" based on the recognition result of the first recognition unit 31, the calculation unit 35 refers to the lane number risk table TW and derives "W1" corresponding to the number of lanes "1 lane" as the lane number risk W. Furthermore, when the calculation unit 35 determines that the number of lanes on the travel road RD is "2 lanes" based on the recognition result of the first recognition unit 31, the calculation unit 35 refers to the lane number risk table TW and derives "W2" corresponding to the number of lanes "2 lanes" as the lane number risk W.

[0120] According to the lane number risk table TW shown in Fig. 8, the greater the number of lanes on the travel route RD, in other words, the greater the road width of the travel route RD, the smaller the value that can be derived as the lane number risk W. This is because it is assumed that on roads with many lanes (in other words, roads with a wide road width), there are fewer moving objects, such as pedestrians, that may unexpectedly enter the road compared to roads with fewer lanes (in other words, roads with a narrow road width).

[0121] Also, here, the number of lanes on the travel route RD is considered to be the road width of the travel route RD, and the lane number risk W corresponding to the number of lanes on the travel route RD is derived, but this is not limited to this. For example, instead of the lane number risk table TW, a table specifying the lane number risk W (in other words, road width risk) corresponding to the road width may be prepared, and the calculation unit 35 may refer to this table to derive the lane number risk W corresponding to the road width of the travel route RD recognized by the first recognition unit 31.

[0122] (Crosswalk risk C) Next, a method for deriving the pedestrian crossing risk C will be described with reference to Fig. 9. The calculation unit 35 derives the pedestrian crossing risk C based on the recognition result of the first recognition unit 31, for example, by referring to a pedestrian crossing risk table TC shown in Fig. 9. Here, the pedestrian crossing risk table TC is a table that defines the pedestrian crossing risk C corresponding to the distance between the obstruction SO and the pedestrian crossing, and is stored in advance in a memory unit of the control device 30 or the like.

[0123] In the pedestrian crossing risk table TC shown in Fig. 9, the distance "less than Dth1" is associated with the pedestrian crossing risk C of "C1," the distance "less than Dth2 but equal to or greater than Dth1" is associated with the pedestrian crossing risk C of "C2," and the distance "equal to or greater than Dth2" is associated with the pedestrian crossing risk C of "C3." Here, C1, C2, and C3 are each predetermined values, and as shown in Fig. 9, there is a relationship of C1>C2>C3.

[0124] For example, if the calculation unit 35 determines, based on the recognition result of the first recognition unit 31, that the distance between the obstructing object SO and the crosswalk closest to that obstructing object SO is "less than Dth1," it refers to the crosswalk risk table TC and derives "C1," which corresponds to the distance "less than Dth1," as the crosswalk risk C. Also, if the calculation unit 35 determines, based on the recognition result of the first recognition unit 31, that the distance between the obstructing object SO and the crosswalk closest to that obstructing object SO is "less than Dth2 but equal to or greater than Dth1," it refers to the crosswalk risk table TC and derives "C2," which corresponds to the distance "less than Dth2 but equal to or greater than Dth1," as the crosswalk risk C.

[0125] According to the crosswalk risk table TC shown in Fig. 9, the shorter the distance between the obstructing object SO and the crosswalk, the larger the value of the crosswalk risk C that can be derived. This is because it is assumed that there are many moving objects, such as pedestrians, that may unexpectedly enter the road near the crosswalk.

[0126] After deriving various parameters such as the obstruction size risk S, guardrail risk G, lane number risk W, and pedestrian crossing risk C, the calculation unit 35 calculates the risk level VR using, for example, the following equation (2). In the following equation (2), α is a gain for the guardrail risk G and is a predetermined value that has been set in advance. Similarly, β is a gain for the pedestrian crossing risk C and is a predetermined value that has been set in advance.

[0127] Risk level VR = Number of lanes risk W × Shield size risk S + (Guardrail risk G × α + Pedestrian crossing risk C × β) (2)

[0128] As described above, the calculation unit 35 can calculate the risk degree VR based on the overlapping area ratio Sr, which is the ratio between the size of the processing target area PT or the size of the non-overlapping area NA, which is the processing target area PT minus the overlapping area OA, and the size of the overlapping area OA. As described above, the overlapping area ratio Sr can also be referred to as the "ratio of the size of the invisible area to the size of the area that should actually be seen." Therefore, by calculating the risk degree VR based on the overlapping area ratio Sr, it is possible to calculate the risk degree VR that takes into account the "ratio of the size of the invisible area to the size of the area that should actually be seen," and it is possible to calculate an appropriate risk degree VR that is suited to the surrounding conditions of the vehicle 1.

[0129] Furthermore, the calculation unit 35 calculates the risk level VR based on the road width (e.g., the number of lanes) of the travel route RD, and can calculate a lower risk level VR when the road width of the travel route RD is large (e.g., when there are many lanes) compared to when the road width is small (e.g., when there are few lanes). This makes it possible to calculate an appropriate risk level VR that takes into account the tendency for fewer moving objects, such as pedestrians, to unexpectedly enter the road on roads with large road widths (e.g., roads with many lanes) compared to roads with small road widths (e.g., roads with few lanes).

[0130] Furthermore, the calculation unit 35 calculates the risk level VR based on an object present in the vicinity of the travel path RD, and when the object is a structure such as a guardrail that prevents entry onto the travel path RD at the boundary between the travel path RD and its surroundings, it can calculate a lower risk level VR compared to when such a structure is not present. This makes it possible to calculate an appropriate risk level VR that takes into account the tendency for fewer moving objects, such as pedestrians, to unexpectedly enter the road on roads where structures such as guardrails that prevent entry onto the road are installed, compared to roads where such structures are not installed.

[0131] Furthermore, the calculation unit 35 calculates the risk level VR based on a crosswalk that exists near the obstructing object SO, and can calculate a higher risk level VR when a crosswalk exists near the obstructing object SO compared to when the crosswalk does not exist. This makes it possible to calculate an appropriate risk level VR that takes into account the tendency for many moving objects, such as pedestrians, to unexpectedly enter the road near crosswalks.

[0132] (2-3. An example of how to identify obstacles blocking an intersection) Next, an example of a method for identifying an intersection obstruction SO1 will be described. The control device 30 (for example, the second recognition unit 32) detects the edge of an object (hereinafter also referred to as "edge EG") present around the vehicle 1 based on the detection value of the external sensor 11, such as the radar 113 or LiDAR. Here, the edge EG can be the outer edge of the object in the left-right direction when viewed from the vehicle 1. Then, based on the detection result of such edge EG, the control device 30 derives an obstruction boundary line BL, which is the boundary of the obstruction area SA caused by the object (in other words, the obstruction SO) present around the vehicle 1.

[0133] For example, as shown in Fig. 10, suppose that an obstruction SOa is present on the front left side of a vehicle 1 traveling on a road RD. In such a case, the control device 30 derives a portion of a virtual line VLa passing through a right edge EGa of the obstruction SOa and a reference point RP, the portion being farther from the reference point RP than the edge EGa, as the obstruction boundary line BLa on the right side of the obstruction area SA caused by the obstruction SOa. The reference point RP may be, for example, the front end of the vehicle 1 and the center in the vehicle width direction, but is not limited thereto, and may be, for example, the mounting position of the radar 113.

[0134] Then, the control device 30 (for example, the discrimination unit 33) refers to map information such as the map information database 24 and determines whether the derived occlusion boundary line BL intersects with an intersection link, which is a link corresponding to the intersection RDx that intersects with the driving path RD ahead of the vehicle 1.

[0135] For example, in the example shown in Fig. 10, link Lk1 and link Lk2 are crossroad links (i.e., links corresponding to crossroad RDx), and the crossroad boundary line BLa on the right side of the crossroad area SA caused by the blocking object SOa intersects with link Lk1, which is a crossroad link. In such a case, the control device 30 determines that the blocking object SOa is a crossroad blocking object SO1 that blocks the crossroad RDx. In Fig. 10, node Nd1 provided between link Lk1 and link Lk2 is a node that represents the intersection of the travel lane RD and the crossroad RDx.

[0136] On the other hand, for example, the shielding boundary line BL (e.g., shielding boundary line BLb) of a shielding object SO located behind the intersection of the travel lane RD and the intersection RDx, such as the shielding object SOb shown in FIG. 10, does not intersect with the intersection link (e.g., link Lk1 and link Lk2). In such a case, the control device 30 determines that the shielding object SOa is not an intersection intersection shielding object SO1 (e.g., a non-intersection intersection shielding object S03). In addition, for example, the control device 30 also determines that a shielding object SO, such as the shielding object SOc shown in FIG. 10, whose shielding boundary line BL (e.g., a shielding boundary line BLc) is blocked by another shielding object SO and does not reach the intersection link, is not an intersection intersection shielding object SO1 (e.g., a non-intersection intersection shielding object S03).

[0137] Here, an example of a method for determining the intersection obstruction SO1 has been described, but the connecting road obstruction SO2 can also be determined in a similar manner. That is, when determining the connecting road obstruction SO2, the connecting road link that is the link corresponding to the connecting road RDc can be used instead of the above intersection link. Furthermore, the determination method described here is merely an example, and other determination methods may be used to determine the attributes of the obstruction SO.

[0138] [3. Processing performed by the control device] Next, a description will be given of an example of processing executed by the control device 30. For example, while the ignition power of the vehicle 1 is on, the control device 30 repeatedly executes a series of processing shown in the flowchart of Fig. 11 at predetermined intervals (for example, every 5 ms).

[0139] 11, first, the control device 30 executes a guardrail risk derivation process (step Sp1) to derive a guardrail risk G, and a lane number risk derivation process (step Sp2) to derive a lane number risk W. An example of a method for deriving the guardrail risk G and the lane number risk W has been described above, so a description thereof will be omitted here.

[0140] Next, the control device 30 performs the following processes for each obstruction SO detected based on the recognition results of the first recognition unit 31: a crosswalk risk derivation process (step Sp3) to derive a crosswalk risk C, an obstruction size risk derivation process (step Sp4) to derive an obstruction size risk S, and a risk degree calculation process (step Sp5) to calculate a risk degree VR.

[0141] An example of a method for deriving the crosswalk risk C has been described above, and therefore its description will be omitted here. An example of the obstruction size risk derivation process in step Sp4 will be described later using Figs. 12 to 14. Furthermore, in the risk degree calculation process in step Sp5, the control device 30 calculates a risk degree VR corresponding to the obstructing object SO that is the current processing target, based on the guardrail risk G, lane number risk W, crosswalk risk C, and obstruction size risk S obtained as processing results from each of the processes from step Sp1 to step Sp4. This allows the control device 30 to calculate a risk degree VR corresponding to each obstructing object SO detected based on the recognition result of the first recognition unit 31.

[0142] (Shielding size risk derivation process) Next, an example of the shielding size risk derivation process in step Sp4 will be described with reference to FIGS.

[0143] 12, in the obstruction size risk derivation process, the control device 30 first determines whether the obstruction SO currently being processed is an oncoming vehicle (step Sp11). If it is determined that the obstruction SO currently being processed is not an oncoming vehicle (step Sp11: NO), the control device 30 calculates a minimum braking time Tmin, which is the braking time required when the vehicle 1 is decelerated from the current vehicle speed VP at a predetermined deceleration x (step Sp12).

[0144] Next, the control device 30 determines whether the current processing target obstacle SO is the intersection obstacle SO1 (step Sp13). If it is determined that the current processing target obstacle SO is the intersection obstacle SO1 (step Sp13: YES), the control device 30 sets a processing target area PT having a lateral distance d1 corresponding to the virtual movement distance Dv, which is the product of the first speed (e.g., 30 km / h) and the minimum braking time Tmin calculated by the processing of step Sp12 (step Sp14).

[0145] Next, the control device 30 derives an overlapping area ratio Sr based on the size of the processing target area PT set in the current occlusion size risk derivation process and the size of the occlusion area SA on the forward image FI caused by the occlusion object SO that is the current processing target (step Sp15).Then, the control device 30 derives an occlusion size risk S corresponding to the occlusion object SO that is the current processing target based on the overlapping area ratio Sr derived by the process of step Sp15, for example, by referring to the occlusion size risk table TS, and ends the current occlusion size risk derivation process.

[0146] Furthermore, in the process of step Sp11, if it is determined that the obstructing object SO currently being processed is an oncoming vehicle (step Sp11: YES), the control device 30 proceeds to the process of step Sp17 shown in FIG.

[0147] Then, the control device 30 determines whether the oncoming vehicle serving as the obstructing object SO to be processed this time is a stopped vehicle (i.e., its traveling speed is equal to or less than a threshold value) (step Sp17). If it is determined that the oncoming vehicle serving as the obstructing object SO to be processed this time is a stopped vehicle (step Sp17: YES), the control device 30 sets a processing target region PT having a lateral distance d1 corresponding to the virtual movement distance Dv, which is the product of the second speed (e.g., 5 km / h) and the minimum braking time Tmin calculated by the processing of step Sp12 (step Sp18), and proceeds to the processing of step Sp15 shown in FIG. 12.

[0148] On the other hand, if it is determined that the oncoming vehicle serving as the obstruction SO currently being processed is not a stopped vehicle (step Sp17: NO), the control device 30 determines whether the inter-vehicle distance between the oncoming vehicle and the following vehicle traveling behind it is equal to or greater than a predetermined value (e.g., 30 m) (step Sp19). Here, the predetermined value can be determined arbitrarily by, for example, the manufacturer of the vehicle 1.

[0149] Then, if it is determined that the inter-vehicle distance between the oncoming vehicle, which is the obstacle SO to be processed this time, and the following vehicle is greater than or equal to a predetermined value (step Sp19: YES), the control device 30 proceeds to the processing of the aforementioned step Sp18 and sets a processing target area PT having a lateral distance d1 according to the second speed and the minimum braking time Tmin.

[0150] On the other hand, if it is determined that the inter-vehicle distance between the oncoming vehicle as the current target obstruction SO and the following vehicle is less than the predetermined value (step Sp19: NO), the risk level VR corresponding to the current target obstruction SO is set to the minimum value (step Sp20), and the current obstruction size risk derivation process is terminated. The minimum value set by the process of step Sp20 is determined in advance by, for example, the manufacturer of the vehicle 1. Furthermore, if the process of step Sp20 is performed, the control device 30 does not need to perform the process of step Sp5 shown in FIG. 11 for the current target obstruction SO.

[0151] By processing step Sp20, if the distance between the oncoming vehicle as an obstruction SO and the following vehicle is short and it is assumed that there is a low possibility that a moving object such as a pedestrian will enter the driving path RD from behind the oncoming vehicle, it is possible to prevent an excessively large value from being calculated as the risk level VR corresponding to the oncoming vehicle.

[0152] Also, in the processing of step Sp13 shown in Figure 12, if it is determined that the obstacle SO currently being processed is not the intersection obstacle SO1 (step Sp13: NO), the control device 30 proceeds to the processing of step Sp21 shown in Figure 14.

[0153] Then, the control device 30 determines whether the current processing target obstacle SO is the connecting road obstacle SO2 (step Sp21). If it is determined that the current processing target obstacle SO is the connecting road obstacle SO2 (step Sp21: YES), the control device 30 sets a processing target area PT having a lateral distance d1 corresponding to the virtual movement distance Dv, which is the product of the third speed (e.g., 15 km / h) and the minimum braking time Tmin calculated by the processing of step Sp12 (step Sp22), and proceeds to the processing of step Sp15 shown in FIG. 12.

[0154] On the other hand, if it is determined that the obstacle SO to be processed this time is not a connecting road obstacle SO2 (step Sp21: NO), that is, if the obstacle SO to be processed this time is a non-intersection road obstacle SO3 other than an oncoming vehicle, the control device 30 sets a processing target area PT having a lateral distance d1 corresponding to the virtual movement distance Dv, which is the product of the second speed (e.g., 5 km / h) and the minimum braking time Tmin calculated by the processing of step Sp12 (step Sp23), and proceeds to processing of step Sp15 shown in Figure 12.

[0155] As described above, the control device 30 determines the attributes of the obstructing object SO and sets the processing target region PT according to the determination result, thereby making it possible to calculate an appropriate risk level VR that takes into account the attributes of the obstructing object SO. Therefore, it becomes possible to calculate an appropriate risk level that is suited to the surrounding conditions of the vehicle 1.

[0156] Furthermore, when an oncoming vehicle facing the opposite direction from vehicle 1 is recognized as an obstruction SO and the inter-vehicle distance between the oncoming vehicle and the following vehicle is equal to or less than a predetermined value, control device 30 can set the risk degree VR corresponding to the oncoming vehicle as the obstruction SO to a settable minimum value. This makes it possible to prevent an excessively large value from being calculated as the risk degree VR corresponding to the oncoming vehicle when it is assumed that there is a low possibility that a moving object such as a pedestrian will enter the driving road RD from behind the oncoming vehicle as the obstruction SO.

[0157] The control method described in this embodiment can be realized by executing a prepared program (control program) on a computer. The control program is, for example, stored in a computer-readable storage medium and executed by being read from the storage medium. The control program may be provided in a form stored in a non-volatile (non-transient) storage medium such as a flash memory, or may be provided via a network such as the Internet. In this embodiment, the computer that executes the control program is the control device 30 (e.g., the processor of the control device 30), but this is not limited thereto. For example, the computer that executes the control program may be included in the vehicle 1, or may be included in an external device 2 that can communicate with the vehicle 1.

[0158] Although one embodiment of the present invention has been described above, it goes without saying that the present invention is not limited to the above-described embodiment. It is clear that a person skilled in the art can conceive of various modifications and alterations within the scope of the claims, and it is understood that these also naturally fall within the technical scope of the present invention.

[0159] For example, even if an intersection obstruction SO1 is present, the control device 30 may set the processing target region PT having a lateral distance d1 corresponding to a speed (e.g., 5 km / h) lower than the first speed and the minimum braking time Tmin, if the distance from the vehicle 1 to the intersection between the intersection RDx blocked by the intersection obstruction SO1 and the travel path RD is equal to or greater than a predetermined value greater than the minimum braking distance Dmin. In this way, when the distance from the vehicle 1 to the intersection is greater than a predetermined value (i.e., when the distance to the intersection is sufficient), the processing target region PT can be set to have a relatively small lateral distance d1, making it possible to prevent an excessively large value from being calculated as the risk level VR.

[0160] Similarly, even if a connecting road obstruction SO2 is present, the control device 30 may set the processing target area PT having a lateral distance d1 corresponding to a speed lower than the third speed (e.g., 5 km / h) and the minimum braking time Tmin, if the distance from the vehicle 1 to the junction point between the connecting road RDc and the travel road RD, which is blocked by the connecting road obstruction SO2, is equal to or greater than a predetermined value greater than the minimum braking distance Dmin. In this way, when the distance from the vehicle 1 to the junction point is greater than a predetermined value (i.e., when the distance to the junction point is sufficient), the processing target area PT can be set to have a relatively small lateral distance d1, making it possible to prevent an excessively large value from being calculated as the risk level VR.

[0161] Furthermore, for example, when a traffic light corresponding to the travel path RD is present at the intersection between the travel path RD and the intersection RDx, the control device 30 may recognize the lighting state of the traffic light using the first recognition unit 31 and calculate the risk degree VR corresponding to the intersection obstruction SO1 that blocks at least a portion of the intersection RDx, taking the lighting state into consideration. More specifically, when a traffic light corresponding to the travel path RD is present at the intersection between the travel path RD and the intersection RDx and the light color of the traffic light indicates that it is possible to proceed, the control device 30 may set the risk degree VR corresponding to the intersection obstruction SO1 that blocks at least a portion of the intersection RDx to a settable minimum value. This makes it possible to prevent an excessively large value from being calculated as the risk degree VR corresponding to the intersection obstruction SO1 that blocks at least a portion of the intersection RDx when it is assumed that there is a low possibility that a moving object will enter the travel path RD from the intersection RDx.

[0162] This specification and the like describe at least the following items. Note that the components in parentheses correspond to those in the above-described embodiments, but are not limited to these.

[0163] (1) A vehicle control device (control device 30) for controlling a vehicle (vehicle 1), a first recognition unit (first recognition unit 31) that recognizes the surrounding situation of the vehicle based on information obtained by an external sensor (external sensor 11) including a camera (camera 111) that captures at least an image in front of the vehicle; a second recognition unit (second recognition unit 32) that recognizes a shielded area (shielded area SA) shielded by an obstacle (shielding object SO) present in front of the vehicle when the first recognition unit recognizes the obstacle; a discrimination unit (discrimination unit 33) that discriminates the attribute of the obstructing object based on the recognition result of the first recognition unit; a setting unit (setting unit 34) that sets the occluded area and a predetermined processing target area (processing target area PT) in a forward image (forward image FI) of the vehicle obtained based on the image pickup result of the camera, based on the recognition result of the second recognition unit and the discrimination result of the discrimination unit; a calculation unit (calculation unit 35) that calculates a risk level (risk level VR) corresponding to the obstruction based on the size of an overlapping area (overlapping area OA) between the obstruction area and the processing target area set by the setting unit; a vehicle control unit (vehicle control unit 36) that controls the vehicle based on the risk degree calculated by the calculation unit; Equipped with The setting unit When the obstacle is determined to be an intersection obstacle (intersection obstacle SO1) that obstructs at least a part of an intersection (intersection RDx) that is a road that intersects with the travel path (travel path RD) on which the vehicle is traveling in front of the vehicle, In the forward image, a region extending toward the intersection obstruction along a first direction corresponding to the width direction of the vehicle is set as the processing target region based on a virtual collision point (virtual collision point CP) between the vehicle and a moving body (virtual moving body VM) that may enter the traveling path from the intersection on the intersection obstruction side at a first speed; The length of the processing target area in the first direction (lateral distance d1 of the processing target area) is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. Vehicle control device.

[0164] According to (1), assuming a configuration in which a risk level corresponding to an obstacle present ahead of the host vehicle is calculated using a forward image of the host vehicle, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the host vehicle. Specifically, when an intersection obstacle that obstructs at least a portion of an intersection that intersects with the travel path on which the host vehicle is traveling and ahead of the host vehicle is present ahead of the host vehicle, the risk level corresponding to the intersection obstacle can be calculated by taking into account a moving object that may enter the travel path at a first speed from the intersection obstructed by the intersection obstacle. Therefore, with a simple configuration, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the vehicle, and it is possible to appropriately control the vehicle based on the risk level. This can ultimately contribute to the development of a sustainable transportation system.

[0165] (2) The vehicle control device according to (1), The setting unit When the obstacle is determined to be a non-intersection obstacle (non-intersection obstacle SO3) different from the intersection obstacle, In the forward image, a region extending toward the non-intersection road obstruction along the first direction is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the road from the obstruction region by the non-intersection road obstruction at a second speed lower than the first speed, and The length of the processing target region in the first direction is set to a length corresponding to a distance determined based on the braking time and the second speed. Vehicle control device.

[0166] According to (2), when a non-intersection road obstruction is present in front of the vehicle, the risk level corresponding to the non-intersection road obstruction can be calculated by taking into account a moving object that may enter the roadway at a second speed from the area obstructed by the non-intersection road obstruction.

[0167] (3) The vehicle control device according to (1) or (2), The setting unit When the obstruction is determined to be a connecting road obstruction (connecting road obstruction SO2) that obstructs at least a part of a connecting road (connecting road RDc) that is a road connected to the travel path in front of the vehicle, In the forward image, a region extending toward the connecting road obstruction along the first direction is set as the processing target region based on a virtual collision point (virtual collision point CP) between the vehicle and a moving body (virtual moving body VM) that may enter the traveling road from the connecting road at a third speed lower than the first speed; The length of the processing target region in the first direction is set to a length corresponding to a distance determined based on the braking time and the third speed. Vehicle control device.

[0168] According to (3), when there is a connecting road obstruction in front of the vehicle that blocks at least a portion of a connecting road, which is a road that connects to the driving road in front of the vehicle, the risk level corresponding to the connecting road obstruction can be calculated by taking into account a moving object that may enter the driving road from the connecting road blocked by the connecting road obstruction at a third speed.

[0169] (4) A vehicle control device according to any one of (1) to (3), the determination unit is configured to be able to refer to map information (map information database 24), and determines the attribute of the obstructing object based on the recognition result of the first recognition unit and the map information; Vehicle control device.

[0170] According to (4), it is possible to determine the attribute of the obstructing object more accurately than when the attribute of the obstructing object is determined based only on the recognition result of the first recognition unit.

[0171] (5) A vehicle control device according to any one of (1) to (4), the setting unit sets the processing target area extending in the first direction and having a predetermined height (height h) in a second direction corresponding to the up-down direction of the vehicle. Vehicle control device.

[0172] According to (5), the setting of the processing target area can be simplified, and the processing load when setting the processing target area can be reduced.

[0173] (6) A vehicle control device according to any one of (1) to (5), The calculation unit calculates the risk degree based on a ratio between the size of the processing target area or the size of an area (non-overlapping area NA) obtained by excluding the overlapping area from the processing target area, and the size of the overlapping area. Vehicle control device.

[0174] According to (6), it is possible to calculate an appropriate degree of risk according to the surrounding circumstances of the vehicle using an image of the front of the vehicle.

[0175] (7) A vehicle control device according to any one of (1) to (6), The calculation unit Calculating the risk level based on the road width of the travel path; When the road width is large, the risk degree is calculated to be smaller than when the road width is small. Vehicle control device.

[0176] According to (7), it is possible to calculate an appropriate degree of risk taking into account the tendency that on roads with large widths (e.g., roads with many lanes), there are fewer pedestrians and other moving objects that unexpectedly enter the road compared to roads with small widths (e.g., roads with few lanes).

[0177] (8) A vehicle control device according to any one of (1) to (7), The calculation unit Calculating the degree of risk based on an object present in the vicinity of the travel path; When a structure that prevents the vehicle from entering the travel path is present at the boundary between the travel path and its outside as the object, the risk degree is calculated to be smaller than when the structure is not present. Vehicle control device.

[0178] According to (8), it is possible to calculate an appropriate level of risk taking into account the tendency that roads equipped with guardrails and other structures that prevent people from entering the roadway have fewer pedestrians and other moving objects that unexpectedly enter the roadway compared to roads without such structures.

[0179] (9) A vehicle control device according to any one of (1) to (8), The calculation unit Calculating the risk level based on a crosswalk that exists near the obstacle; When the crosswalk is present near the obstacle, the degree of risk is calculated to be greater than when the crosswalk is not present. Vehicle control device.

[0180] According to (9), it is possible to calculate an appropriate degree of risk taking into account the tendency for many pedestrians and other moving objects to unexpectedly enter the road near crosswalks.

[0181] (10) A vehicle control device according to any one of (1) to (9), The vehicle control device includes: When another vehicle facing in the opposite direction to the vehicle is recognized as the obstructing object and the inter-vehicle distance between the other vehicle and the following vehicle is equal to or less than a predetermined value, the risk degree corresponding to the other vehicle as the obstructing object is set to a settable minimum value; Vehicle control device.

[0182] According to (10), when the distance between the other vehicle acting as an obstruction and the following vehicle is short and it is assumed that there is a low possibility that a moving object such as a pedestrian will enter the roadway from behind the other vehicle, it is possible to prevent an excessively large value from being calculated as the risk level corresponding to the other vehicle.

[0183] (11) A vehicle control device according to any one of (1) to (10), The vehicle control device includes: If a traffic light corresponding to the travel path is present at the intersection between the travel path and the intersection, and the light color of the traffic light indicates that it is possible to proceed, the risk degree corresponding to the intersection obstruction is set to a settable minimum value. Vehicle control device.

[0184] According to (11), when it is assumed that there is a low possibility of a moving object entering the travel lane from an intersection, it is possible to prevent an excessively large value from being calculated as the risk level corresponding to an intersection obstruction that blocks at least part of the intersection.

[0185] (12) A computer (control device 30) that controls a vehicle (vehicle 1) Recognizing the surrounding situation of the vehicle based on information obtained by an external sensor (external sensor 11) including a camera (camera 111) that captures at least an image in front of the vehicle; When an obstruction (obstruction SO) present in front of the vehicle is recognized, an obstruction area (obstruction area SA) obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; Based on the recognition result of the obstructed area and the determination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area are set for a forward image (forward image FI) of the vehicle obtained based on the imaging result of the camera; Calculating a risk level (risk level VR) corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Processing is performed, In the process of setting the processing target area, When the obstacle is determined to be an intersection obstacle (intersection obstacle SO1) that obstructs at least a part of an intersection (intersection RDx) that is a road that intersects with the travel path (travel path RD) on which the vehicle is traveling in front of the vehicle, In the forward image, a region extending toward the intersection obstruction along a first direction corresponding to the width direction of the vehicle is set as the processing target region based on a virtual collision point (virtual collision point CP) between the vehicle and a moving body (virtual moving body VM) that may enter the traveling path from the intersection on the intersection obstruction side at a first speed; The length of the processing target area in the first direction (lateral distance d1 of the processing target area) is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. Control method.

[0186] According to (12), assuming a configuration in which a risk level corresponding to an obstacle present ahead of the host vehicle is calculated using a forward image of the host vehicle, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the host vehicle. Specifically, when an intersection obstacle that obstructs at least a portion of an intersection that intersects with the travel path on which the host vehicle is traveling and ahead of the host vehicle is present ahead of the host vehicle, the risk level corresponding to the intersection obstacle can be calculated by taking into account a moving object that may enter the travel path at a first speed from the intersection obstructed by the intersection obstacle. Therefore, with a simple configuration, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the vehicle, and it is possible to appropriately control the vehicle based on the risk level. This can ultimately contribute to the development of a sustainable transportation system.

[0187] (13) A computer (control device 30) that controls a vehicle (vehicle 1) Recognizing the surrounding situation of the vehicle based on information obtained by an external sensor (external sensor 11) including a camera (camera 111) that captures at least an image in front of the vehicle; When an obstruction (obstruction SO) present in front of the vehicle is recognized, an obstruction area (obstruction area SA) obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; Based on the recognition result of the obstructed area and the discrimination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area (processing target area PT) are set for a front image of the vehicle (front image FI) obtained based on the imaging result of the camera; Calculating a risk level (risk level VR) corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Let the processing take place, In the process of setting the processing target area, When the obstacle is determined to be an intersection obstacle (intersection obstacle SO1) that obstructs at least a part of an intersection (intersection RDx) that is a road that intersects with the travel path (travel path RD) on which the vehicle is traveling in front of the vehicle, In the forward image, a region extending toward the intersection obstruction along a first direction corresponding to the width direction of the vehicle is set as the processing target region based on a virtual collision point (virtual collision point CP) between the vehicle and a moving body (virtual moving body VM) that may enter the traveling path from the intersection on the intersection obstruction side at a first speed; The length of the processing target area in the first direction (lateral distance d1 of the processing target area) is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed. Control program.

[0188] According to (13), assuming a configuration in which a risk level corresponding to an obstacle present ahead of the host vehicle is calculated using a forward image of the host vehicle, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the host vehicle. Specifically, when an intersection obstacle that obstructs at least a portion of an intersection that intersects with the travel path on which the host vehicle is traveling and ahead of the host vehicle is present ahead of the host vehicle, the risk level corresponding to the intersection obstacle can be calculated by taking into account a moving object that may enter the travel path at a first speed from the intersection obstructed by the intersection obstacle. Therefore, with a simple configuration, it is possible to calculate an appropriate risk level in accordance with the surrounding conditions of the vehicle, and it is possible to appropriately control the vehicle based on the risk level. This can ultimately contribute to the development of a sustainable transportation system. [Explanation of symbols]

[0189] 1 vehicle 30 Control device (vehicle control device) 11 External Sensors 31 1st recognition part 32 Second recognition part 33 Discrimination part 34 Setting section 35 Calculation section 36 Vehicle control unit 111 Camera CP Virtual Collision Point d1 Horizontal distance of the processing area FI forward image h Height of the area to be processed NA Non-overlapping area OA overlap area PT Processing Area RD driving route RDx Crossroads SA shielding area SO Cover SO1 Crossroads Cover SO2 Connection Shield VM Virtual Machine

Claims

1. A vehicle control device that controls a vehicle, a first recognition unit that recognizes a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; a second recognition unit that recognizes an obstruction area obstructed by the obstruction when the first recognition unit recognizes an obstruction present in front of the vehicle; a discrimination unit that discriminates an attribute of the obstructing object based on a recognition result of the first recognition unit; a setting unit that sets the obstructed area and a predetermined processing target area in a forward image of the vehicle obtained based on an image capturing result of the camera, based on a recognition result of the second recognition unit and a discrimination result of the discrimination unit; a calculation unit that calculates a risk level corresponding to the obstructing object based on the size of an overlapping area between the obstructed area set by the setting unit and the processing target area; a vehicle control unit that controls the vehicle based on the risk degree calculated by the calculation unit; Equipped with The setting unit When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and a length in the first direction of the processing target region is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed; Vehicle control device.

2. The vehicle control device according to claim 1, The setting unit When the obstacle is determined to be a non-intersection obstacle different from the intersection obstacle, In the forward image, a region extending toward the non-intersection road obstruction along the first direction is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the road from the obstruction region due to the non-intersection road obstruction at a second speed lower than the first speed, and The length of the processing target region in the first direction is set to a length corresponding to a distance determined based on the braking time and the second speed. Vehicle control device.

3. The vehicle control device according to claim 1, The setting unit When the obstruction is determined to be a connecting road obstruction that obstructs at least a part of a connecting road that is a road connected to the travel road ahead of the vehicle, In the forward image, a region extending toward the connecting road obstruction along the first direction is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the traveling road from the connecting road at a third speed lower than the first speed, and a length in the first direction of the processing target region is set to a length corresponding to a distance determined based on the braking time and the third speed; Vehicle control device.

4. The vehicle control device according to claim 1, the determination unit is configured to be able to refer to map information, and determines an attribute of the obstructing object based on the recognition result of the first recognition unit and the map information. Vehicle control device.

5. 5. A vehicle control device according to claim 1, the setting unit sets the processing target area extending in the first direction and having a predetermined height in a second direction corresponding to a vertical direction of the vehicle. Vehicle control device.

6. 5. A vehicle control device according to claim 1, the calculation unit calculates the degree of risk based on a ratio between a size of the processing target area or a size of an area obtained by excluding the overlapping area from the processing target area, and a size of the overlapping area; Vehicle control device.

7. 5. A vehicle control device according to claim 1, The calculation unit Calculating the risk level based on the road width of the travel path; When the road width is large, the risk degree is calculated to be smaller than when the road width is small. Vehicle control device.

8. 5. A vehicle control device according to claim 1, The calculation unit Calculating the risk level based on an object present in the vicinity of the travel path; When a structure that prevents the vehicle from entering the travel path is present at the boundary between the travel path and its outside as the object, the risk degree is calculated to be smaller than when the structure is not present. Vehicle control device.

9. 5. A vehicle control device according to claim 1, The calculation unit Calculating the risk level based on a crosswalk that exists near the obstacle; When the crosswalk is present near the obstacle, the degree of risk is calculated to be greater than when the crosswalk is not present. Vehicle control device.

10. 5. A vehicle control device according to claim 1, The vehicle control device includes: When another vehicle facing in the opposite direction to the vehicle is recognized as the obstructing object, and the inter-vehicle distance between the other vehicle and the following vehicle is equal to or less than a predetermined value, the risk degree corresponding to the other vehicle as the obstructing object is set to a settable minimum value; Vehicle control device.

11. 5. A vehicle control device according to claim 1, The vehicle control device includes: If a traffic light corresponding to the travel path is present at the intersection between the travel path and the intersection, and the light color of the traffic light indicates that it is possible to proceed, the risk degree corresponding to the intersection obstruction is set to a settable minimum value. Vehicle control device.

12. The computer that controls the vehicle Recognizing a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; When an obstruction present in front of the vehicle is recognized, an obstruction area obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; based on the recognition result of the obstructed area and the determination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area are set for the image of the front of the vehicle obtained based on the image pickup result of the camera; Calculating a risk level corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Processing is performed, In the process of setting the processing target area, When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and a length in the first direction of the processing target region is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed; Control method.

13. The computer that controls the vehicle Recognizing a surrounding situation of the vehicle based on information obtained by an external sensor including a camera that captures at least an image in front of the vehicle; When an obstruction present in front of the vehicle is recognized, an obstruction area obstructed by the obstruction is recognized; determining attributes of the obstructing object based on the result of the recognition of the surrounding situation; based on the recognition result of the obstructed area and the determination result of the attribute of the obstructing object, the obstructed area and a predetermined processing target area are set for the image of the front of the vehicle obtained based on the image pickup result of the camera; Calculating a risk level corresponding to the obstruction based on the size of an overlapping area between the set obstruction area and the processing target area; controlling the vehicle based on the degree of risk; Let the processing take place, In the process of setting the processing target area, When the obstruction is determined to be an intersection obstruction that obstructs at least a part of an intersection that is a road that intersects the road on which the vehicle is traveling ahead of the vehicle, In the forward image, a region extending toward the intersection obstruction side along a first direction corresponding to a width direction of the vehicle is set as the processing target region based on a virtual collision point between the vehicle and a moving object that may enter the travel path from the intersection on the intersection obstruction side at a first speed, and a length in the first direction of the processing target region is set to a length corresponding to a braking time when the vehicle is decelerated at a predetermined deceleration and a distance according to the first speed; Control program.

Images