JP2020129331A - Reverse run determination system, reverse run determination method, and reverse run determination program - Google Patents

Abstract

To provide a technique for minimizing a possibility of erroneous determination when determining a possibility of reverse run by a vehicle intending to enter a road from an outer side in a road width direction.SOLUTION: A reverse run determination system 1 includes: a feature recognition section 11 for recognizing a relative angle of a vehicle body longitudinal direction of an object vehicle and an extending direction of a separation feature and also a relative distance between the object vehicle and the separation feature when the object vehicle is intending to enter an object road from an outer side on a first side in a width direction; and a determination section 12 for determining whether or not the object vehicle can perform forward run entrance for transferring a state into a forward traveling state after entering an object lane only by forward traveling based on the relative angle and the relative distance, and determining that there is a possibility of reverse run on the object lane by the object vehicle when it is determined that the object vehicle cannot perform forward run entrance.SELECTED DRAWING: Figure 1

Description

The present invention relates to a reverse running determination system, a reverse running determining method, and a reverse run determining program for determining the possibility that a vehicle entering a road from the outside in the road width direction will run backward.

An example of a system for detecting reverse running of a vehicle is disclosed in Japanese Patent No. 5044436 (Patent Document 1). The system described in Patent Document 1 targets a vehicle that joins the main line from a merging passage on a highway or the like to detect reverse running. Then, the system described in Patent Document 1 acquires a displacement pattern of a stationary object that changes its position on the image data according to the movement of the vehicle from an image captured by the camera around the vehicle, and the acquired displacement pattern of the stationary object. It is configured to determine whether the vehicle is going to reverse the main line based on the. As a specific example, in Patent Document 1, an angle between a roadside line (an example of a stationary object) and a horizontal axis in image data is acquired, and when the angle decreases to zero or exceeds zero. It is described that it is determined that the vehicle has started reverse running when a predetermined value is reached (paragraphs 0034-0037). In this case, like the vehicle at the position of "V13" in FIG. 3 of Patent Document 1, the direction of the vehicle (direction toward the front, the same applies hereinafter) is parallel to the road width direction of the main line from the state along the traveling direction of the main line. The state of the vehicle changes from a state along the traveling direction of the main line to a state facing the traveling direction of the main line, as in the vehicle at the position of "V14" in FIG. 3 of Patent Document 1. At that point, it is determined that the vehicle has started reverse running.

Japanese Patent No. 5044436

By the way, the vehicle entering from the outside in the width direction of the road is not limited to the confluence point as described in Patent Document 1, but also a point or an intersection (for example, an intersection where the facility facing the road enters the road). , T-junction) and other places other than the confluence. Since reverse running of the vehicle may occur at a place other than these confluence points, it is considered to use the reverse running judgment based on the direction of the vehicle to detect the reverse run at a place other than these confluence points. To be However, even if the driver is performing a normal driving operation at a point or an intersection where the facility enters the road, the shape of the approach road to the destination road (for example, the parking lot exit Due to the shape, the direction of the vehicle when entering the approach road may have a component opposite to the traveling direction. In addition, even when the driver is performing normal driving operation, including the case where the place where the approach road is approached is the confluence, the presence of obstacles and the position of the vehicle before the approach Due to (for example, the parking position), the direction of the vehicle when entering the approach road may have a component opposite to the traveling direction. Therefore, there is a possibility that the rate of erroneously determining that the vehicle may run in reverse may increase even if the driver is performing a normal driving operation, based only on the direction of the vehicle.

Therefore, in the case of determining the possibility that a vehicle that enters the road from the outside in the width direction of the road may run backward, it is desired to realize a technique that can reduce the possibility of erroneous determination.

In the reverse running determination system according to the present disclosure, a road having a separated feature that is a feature that separates lanes having different traveling directions from each other is set as a target road, and one of the road widthwise directions with respect to the separated feature on the target road is defined. When the target vehicle is approaching the target road from the outside on the first side in the width direction with the lane on the first side in the width direction as the target lane, the vehicle is separated from the vehicle body front-back direction of the target vehicle. A feature recognition unit that recognizes a relative angle with respect to the extension direction of the feature and a relative distance between the target vehicle and the separated feature, and travels in the forward direction by entering the target lane only by traveling forward. It is determined based on the relative angle and the relative distance whether or not the target vehicle can perform a forward approach to shift to a state in which the target vehicle cannot perform the forward approach. In this case, the determination unit determines that the target vehicle may run backward in the target lane.

In the reverse running determination method according to the present disclosure, a road having a separated feature that is a feature that separates lanes having different traveling directions from each other is set as a target road, and one of the road widthwise directions with respect to the separated feature on the target road is used. When the target vehicle is approaching the target road from the outside on the first side in the width direction with the lane on the first side in the width direction as the target lane, the vehicle is separated from the vehicle body front-back direction of the target vehicle. A feature recognition step of recognizing a relative angle with respect to the extension direction of the feature and a relative distance between the target vehicle and the separated feature, and entering the target lane only by traveling forward and traveling in the forward direction It is determined based on the relative angle and the relative distance whether or not the target vehicle can perform a forward approach to shift to a state in which the target vehicle cannot perform the forward approach. In this case, the target vehicle may reverse in the target lane.

The reverse running determination program according to the present disclosure sets a road having a separated feature, which is a feature that separates lanes having different traveling directions, as a target road, and selects one of the road widthwise directions with respect to the separated feature on the target road. When the target vehicle is approaching the target road from the outside on the first side in the width direction with the lane on the first side in the width direction as the target lane, the vehicle is separated from the vehicle body front-back direction of the target vehicle. A feature recognition function for recognizing the relative angle of the extension direction of the feature and the relative distance between the target vehicle and the separated feature, and entering the target lane only by traveling forward and traveling in the forward direction It is determined based on the relative angle and the relative distance whether or not the target vehicle can perform a forward approach to shift to a state in which the target vehicle cannot perform the forward approach. And a determination function of determining that the target vehicle may reversely run in the target lane in the case of being performed.

Depending on the direction and position of the target vehicle when entering the target road, the travel line that the target vehicle can take after entering the target road is determined, so whether or not the target vehicle can make a forward approach, The determination can be made based on the direction and position of the target vehicle when entering the target road. According to the above configuration, the relative angle and the relative distance indicating the direction and the position of the target vehicle when entering the target road are acquired with the separated feature provided on the target road as a reference, and the relative angle and the relative distance are acquired. Based on the distance, it can be determined whether or not the target vehicle can make a forward approach.

Then, in the above configuration, when it is determined that the target vehicle cannot make a forward approach, it is determined that the target vehicle may run backward in the target lane. When the target vehicle cannot make a forward approach, the target vehicle that has entered the target road is likely to reverse the target lane. Therefore, according to the above configuration, the target vehicle is the target lane. It is possible to reduce the possibility that an erroneous determination that does not determine that there is a possibility of reverse running will occur in a situation where there is a high possibility that the vehicle will reverse. On the other hand, in the above configuration, when the target vehicle can make a forward approach, it is not determined that the target vehicle may reverse in the target lane, so that the driver can drive normally. It is also possible to reduce the possibility of making an erroneous determination that there is a possibility that the vehicle may run in reverse while performing an operation.

As described above, according to the above configuration, in the case of determining the possibility that the target vehicle trying to enter the target road from the outer side on the first side in the width direction may run backward, an erroneous determination may occur. It is possible to keep it low.

Further features and advantages of the reverse run determination system, the reverse run determination method, and the reverse run determination program will become more apparent from the following description of the embodiments described with reference to the drawings.

Block diagram showing the schematic configuration of the reverse running determination system Explanatory drawing of each parameter used in the judgment by the judgment unit Diagram showing an example of a situation where the target vehicle makes a forward approach A diagram showing an example of a situation in which the target vehicle runs in the target lane in the opposite direction Flowchart showing an example of the procedure of reverse running determination processing

An embodiment of a reverse running determination system will be described with reference to the drawings. As shown in FIG. 1, the reverse running determination system 1 includes a control unit 10. Although illustration is omitted, the control unit 10 includes an arithmetic processing device such as a single or a plurality of CPUs (Central Processing Units) as a core member and a storage device to which the arithmetic processing device can refer. The storage device is, for example, a RAM (Random Access Memory) or a ROM (Read Only Memory). The control unit 10 executes the respective programs stored in the main storage device (a storage device included in the control unit 10) and the storage device 3 (a storage device provided separately from the control unit 10) to perform the reverse running determination. Realize each function to perform. The storage device 3 includes, as a hardware configuration, a storage medium that can store and rewrite information, such as a flash memory or a hard disk. Note that, in the following description, at least a part of data (including programs) described as being stored in the main storage device is stored in the storage device 3, and when stored in the storage device 3 in the following description. At least a part of the data (including programs) to be described may be stored in the main storage device. In this embodiment, the control unit 10 functions as a “computer”.

As shown in FIG. 1, the reverse running determination system 1 (control unit 10) includes a plurality of functional units. Specifically, the reverse running determination system 1 (control unit 10) includes a feature recognition unit 11 and a determination unit 12, and further includes a position information acquisition unit 13 and a warning processing unit 14 in the present embodiment. There is. The plurality of functional units are configured to exchange information with each other and to extract data from a storage device (main storage device or storage device 3). The plurality of functional units are at least logically distinguished and need not be physically distinguished. Further, the plurality of functional units do not have to be implemented by common hardware, and may be implemented by a plurality of hardware capable of communicating with each other (for example, the control unit 10 and the server device).

The control unit 10 executes the programs stored in the storage device (main storage device or storage device 3) (the programs that form the reverse running determination program) to perform the functions of the functional units illustrated in FIG. Realize. That is, each functional unit of the reverse run determination system 1 is configured by software (program) stored in the storage device, hardware such as a separately provided arithmetic circuit, or both of them. In other words, a program (reverse running determination program) for causing the arithmetic processing device (computer) to realize the function of each functional unit of the reverse running determination system 1 is stored in a storage device that can be referred to by the arithmetic processing device. The reverse running determination program is provided, for example, by a storage medium or via a communication network. When the reverse running determination system 1 is used by being incorporated in a vehicle navigation system, the provided reverse running determination program is installed in, for example, an in-vehicle device of the navigation system, and the reverse running determination system 1 is realized.

As shown in FIG. 1, the reverse running determination system 1 is realized using a plurality of devices (specifically, the devices included in the information acquisition unit 20, the storage device 3, and the warning device 4). The information acquisition unit 20 includes various devices (various sensors) that acquire information necessary for performing reverse running determination. Each device (each sensor) included in the information acquisition unit 20 is, for example, every prescribed cycle (every sampling period) or every time the target vehicle 2 travels a prescribed distance, or the target vehicle 2 reaches a prescribed position. The information is acquired at the timing. The control unit 10 is configured to be able to acquire the information acquired by the information acquisition unit 20 (detection information of each device included in the information acquisition unit 20). In the present specification, the “device” is not limited to one piece of integrated hardware, and one “device” may be composed of a plurality of pieces of hardware (device group) separated from each other. .. In addition, at least a part (for example, the storage device 3) of the plurality of devices for realizing the reverse running determination system 1 is not the target vehicle 2 but a portable device that can be carried by the user (for example, a portable navigation device or a smartphone). Etc.) or an external device (for example, a server device) that can communicate with the target vehicle 2. That is, the reverse running determination system 1 is not limited to the in-vehicle device, and may be realized using a portable device or an external device.

In this embodiment, the information acquisition unit 20 includes a GPS receiver 21, a vehicle speed sensor 22, a gyro sensor 23, and a camera 24. Each of these devices included in the information acquisition unit 20 is provided in the target vehicle 2. The information acquisition unit 20 may further include an acceleration sensor that generates a signal according to the acceleration of the target vehicle 2.

The GPS receiver 21 receives a GPS signal from a GPS (Global Positioning System) satellite and controls a signal for calculating the current position (latitude and longitude) of the target vehicle 2 (vehicle subject to reverse running determination). Output to the unit 10. The vehicle speed sensor 22 generates a signal according to the rotation of the wheels of the target vehicle 2 and sends a signal (for example, a vehicle speed pulse signal) for calculating the traveling speed (vehicle speed) and the traveling distance of the target vehicle 2 to the control unit 10. Output. The gyro sensor 23 generates a signal according to a change in the direction of the target vehicle 2 (the direction toward the front along the vehicle body front-rear direction D of the target vehicle 2) and controls the signal for calculating the yaw rate of the target vehicle 2. Output to the unit 10.

The camera 24 is an imaging device (for example, a front camera) that captures at least the front of the target vehicle 2. That is, the image 7 captured by the camera 24 includes an image (landscape image) in front of the target vehicle 2. The camera 24 generates an image signal at least in front of the target vehicle 2 and outputs the generated image signal to the control unit 10. As schematically shown in the insets of FIG. 3 and FIG. 4 as an example of the image 7 taken by the camera 24, the camera 24 is configured so that the road surface in front of the target vehicle 2 is taken (attachment position, attachment angle, The angle of view, etc.) is set in the target vehicle 2. Further, in the present embodiment, as shown in FIGS. 3 and 4, the camera 24 is installed in the target vehicle 2 so as to have a visual field F that spreads left and right around the front of the target vehicle 2 in the vehicle body front-rear direction D. It The camera 24 is installed, for example, at the center position of the target vehicle 2 in the width direction (horizontal direction H). Here, the left-right direction H is a left-right direction with respect to the direction of the target vehicle 2 (direction orthogonal to the vehicle body front-rear direction D of the target vehicle 2 on the horizontal plane), and changes in accordance with the change of the direction of the target vehicle 2. .. The camera 24 is attached to, for example, the upper part of the front window or the front bumper.

The storage device 3 stores map information. The map information includes road information and feature information. The road information is information on the road network, and includes information on nodes corresponding to intersections and information on links connecting the nodes. Links may be maintained for each lane. The road information includes, for example, information on the node position and attributes (for example, intersection type) as the node information. Further, in the road information, as link information, for example, positions of end points (start point and end point) of the link, link length, number of lanes, road width, lane width, traveling direction M for each lane (determined by traffic rules It includes information on the direction of travel). The feature information is information on various features (for example, buildings, road markings, road signs, traffic lights, etc.) provided on or around the road. The feature information includes, for example, information on the position and attribute of the feature (type, name, etc. of the feature). The storage device 3 stores, as the feature information, information on the separated features 6 to be described later.

A road having a separated feature 6 that is a feature that separates lanes 50 having different traveling directions M is a target road 5, and a width direction that is one side of a road width direction W with respect to the separated feature 6 on the target road 5. With the lane 50 of the first side W1 as the target lane 51, the reverse running determination system 1 determines that the target vehicle 2 is trying to enter the target road 2 from outside the width direction first side W1. This is a system for determining the possibility of traveling backward in the target lane 51. The reverse running determination system 1 can reduce the possibility of making an erroneous decision that does not determine that the target vehicle 2 may run in reverse in a situation where the target vehicle 2 is likely to run backward in the target lane 51. In addition, it is possible to reduce the possibility of erroneous determination that the driver may reverse when the driver is performing a normal driving operation. Hereinafter, the configuration of each functional unit (see FIG. 1) of the reverse running determination system 1 for realizing such a configuration will be specifically described. In addition, as described below, in the present embodiment, the reverse running determination system 1 determines that the target vehicle 2 (the occupant of the target vehicle 2 is occupant when it is determined that the target vehicle 2 may reversely run in the target lane 51. , Especially the driver).

The position information acquisition unit 13 is a functional unit that acquires the current position of the target vehicle 2. The position information acquisition unit 13 acquires the current position of the target vehicle 2 based on the detection results of the GPS receiver 21, the vehicle speed sensor 22, and the gyro sensor 23, for example. At this time, a correction (map matching process) for matching the current position of the target vehicle 2 with the position on the road may be performed based on the road information stored in the storage device 3. Further, the position information acquisition unit 13 may acquire the current position of the target vehicle 2 based on the recognition result of the features (road markings, etc.) around the target vehicle 2. Specifically, the feature around the target vehicle 2 is recognized by image processing of the captured image 7 by the camera 24, and the position information of the feature stored in the storage device 3 and the feature are recognized. The current position of the target vehicle 2 can be acquired based on the relative position of the target vehicle 2 with respect to the feature at the time. Further, the position information acquisition unit 13 acquires the position acquired based on the recognition result of the features (road markings, etc.) around the target vehicle 2 and the traveling locus (transition of the current position) of the target vehicle 2 from the position. Based on this, the current position of the target vehicle 2 may be acquired. The position information acquisition unit 13 executes a process (position information acquisition process) of acquiring the current position of the target vehicle 2, and a function (position information acquisition function) of acquiring the current position of the target vehicle 2 is executed by executing the process. Will be realized.

When the target vehicle 2 is approaching the target road 5 from outside the width-direction first side W1, the feature recognition unit 11 determines the vehicle body longitudinal direction D of the target vehicle 2 and the extending direction of the separated features 6. It is a functional unit that recognizes a relative angle θ1 with respect to E and a first relative distance L1 between the target vehicle 2 and the separated feature 6 (see FIG. 2 ). The feature recognition unit 11 acquires information on the separated feature 6 used for recognizing the relative angle θ1 and the first relative distance L1, for example, by referring to the storage device 3 or by communication such as road-vehicle communication. get. Here, the road-to-vehicle communication is communication between a communication device provided in the target vehicle 2 and a communication device (roadside communication device such as an optical beacon) installed on the road side. In the present embodiment, the first relative distance L1 corresponds to the "relative distance". Further, in the present embodiment, the process executed by the feature recognition unit 11 corresponds to the “feature recognition step”, and the function realized by executing the process corresponds to the “feature recognition function”.

As shown in FIGS. 3 and 4, the separated features 6 are features that divide lanes 50 having different traveling directions M (lanes 50 having opposite traveling directions M). Therefore, the lane 50 on the width-direction second side W2, which is the side opposite to the width-direction first side W1 in the road width direction W with respect to the separated feature 6, faces the target lane 51 with the traveling direction M being the opposite direction. Lane 52. The separated features 6 are continuously or intermittently provided along the boundary between the lanes 50 having different traveling directions M. The separated features 6 may be three-dimensional features or two-dimensional features. For example, the separated features 6 are formed by a three-dimensional structure such as a separation strip (central separation strip) or a road tack, or by road markings (center line, etc.) provided on the road surface with paint or the like. The separating strip is a three-dimensional structure made of, for example, concrete, guardrails, tree planting, or the like. The target road 5 may be limited to a road having a three-dimensional feature as the separated feature 6. Furthermore, the target road 5 is limited to a road having a three-dimensional feature (such as a separation zone) that physically restricts the movement of the target vehicle 2 to the second side W2 in the width direction as the separated feature 6. Good.

The control unit 10 (for example, the feature recognition unit 11) acquires, by the position information acquisition unit 13, whether or not the target vehicle 2 is approaching the target road 5 from outside the width-direction first side W1. Information on the current position of the target vehicle 2 (for example, transition of the current position), map information (road information and feature information) acquired by referring to the storage device 3 or by communication such as road-to-vehicle communication, It judges based on. In determining whether or not the target vehicle 2 is approaching the target road 5 from the outside of the width-direction first side W1, the storage device 3 is referred to, or instead of the map information acquired by communication, or In addition to the map information, the image recognition result (image recognition result of roads and features around the target vehicle 2) for the captured image 7 by the camera 24 may be used.

The target vehicle 2 may enter the target road 5 from the outside in the road width direction W, for example, at a point where the facility that faces the target road 5 enters the target road 5 (e.g., an exit of a gas station or a parking lot). Exit, etc.), a T-shaped road (T-shaped road corresponding to the T-shaped horizontal bar of the T-shaped road) provided in the middle of the target road 5, and a confluence point where another road merges with the target road 5. Etc. In the specific examples shown in FIGS. 3 and 4 which will be described later, the target vehicle 2 is directed to the target road 5 at a point where the target road 5 is approached from a facility facing the target road 5 (here, a general road). It is assumed that the vehicle enters from the outer side of the width direction first side W1.

In the present embodiment, the feature recognition unit 11 acquires the captured image 7 in front of the target vehicle 2 by the camera 24, and performs image recognition processing of the separated features 6 included in the captured image 7 (for example, pattern matching processing or the like). )I do. Then, the feature recognition unit 11 recognizes the relative angle θ1 (see FIG. 2) based on the inclination angle θ2 (see FIGS. 3 and 4) in the extending direction E of the separated feature 6 in the captured image 7. As shown in FIGS. 3 and 4, in the present embodiment, the inclination angle θ2 is the inclination angle with respect to the left-right direction H. In the example shown in FIGS. 3 and 4, it is assumed that the separated feature 6 is a three-dimensional structure protruding upward V with respect to the road surface. In such a case, the inclination angle of the extension direction E of the portions of the separated features 6 having the same height (in the examples shown in FIGS. 3 and 4, the portions of the separated features 6 that are connected to the road surface) is defined as the inclination angle. It can be set to θ2. In addition, when the separated features 6 are intermittently provided along the boundary between the lanes 50 having different traveling directions M, the inclination angle of the virtual straight line passing through each separated feature 6 in the captured image 7 is changed to the inclination angle. It is good to set to θ2.

The feature recognition unit 11 derives the relative angle θ1 from the inclination angle θ2 based on the correspondence between each position (pixel position) in the captured image 7 and the actual position on the road surface (relative position with respect to the target vehicle 2). To do. Specifically, the feature recognition unit 11 performs, for example, projective conversion of the captured image 7 into an image in plan view based on the installation form of the camera 24 on the target vehicle 2, and the position and actual position in the captured image 7. The relative angle θ1 is derived from the inclination angle θ2 by referring to a conversion table that defines the relationship with the position on the road surface. As shown in FIG. 2, in the present embodiment, a first vector A1 that is a vector that is directed forward of the target vehicle 2 along the vehicle body front-rear direction D, and a target lane 51 along the extension direction E of the separated features 6. The angle formed by the second vector A2, which is the vector moving in the traveling direction M, is defined as the relative angle θ1. Therefore, when the relative angle θ1 is larger than 90 degrees, the direction of the target vehicle 2 when entering the target road 5 has a component opposite to the traveling direction M of the target lane 51.

Here, the left side is defined as the first side H1 when the left side traffic is defined by the traffic rules, and the left side is defined as the first side H1 when the right side traffic is defined by the traffic rules. As shown in FIGS. 3 and 4, in the present embodiment, as an example, it is assumed that the traffic rules define left-hand traffic, so the first side H1 is the right side and the second side H2 is On the left. As shown in FIGS. 3 and 4, when the target vehicle 2 enters the target road 5 and the relative angle θ1 (see FIG. 2) is larger than 90 degrees, the extension of the separated features 6 in the captured image 7 is extended. The direction E is inclined toward the upper side V toward the first side H1. Although illustration is omitted, when the relative angle θ1 is 90 degrees when the target vehicle 2 enters the target road 5, the extending direction E of the separated features 6 in the captured image 7 is along the left-right direction H. When the target vehicle 2 enters the target road 5 and the relative angle θ1 is smaller than 90 degrees, the extending direction E of the separated features 6 in the captured image 7 increases toward the second side H2. Incline toward V.

Further, in the present embodiment, the feature recognition unit 11 recognizes the first relative distance L1 based on the image recognition result of the separated feature 6 included in the image 7 captured by the camera 24. As shown in FIG. 2, in the present embodiment, the first relative distance L1 is the relative distance between the reference position in the target vehicle 2 and the separated feature 6 along the road width direction W of the target road 5. In the example shown in FIG. 2, the reference position in the target vehicle 2 is, for example, the center position in the left-right direction H in the front part of the target vehicle 2. The feature recognition unit 11 performs, for example, projective conversion of the captured image 7 into an image in plan view based on the installation form of the camera 24 on the target vehicle 2, and the position in the captured image 7 and the actual position on the road surface. The first relative distance L1 is derived by, for example, referring to a conversion table in which the relationship with At this time, the relative distance between the reference position in the target vehicle 2 and the separated feature 6 along the vehicle body front-rear direction D is derived, and then the first based on the relative distance and the relative angle θ1 (or the inclination angle θ2). The relative distance L1 may be derived. When the separated feature 6 is a three-dimensional structure protruding upward V with respect to the road surface, as in the example shown in FIGS. 3 and 4, for example, a connection portion of the separated feature 6 with the road surface is used. It is preferable to derive the first relative distance L1 by detecting, or using height information of the separated features 6.

The feature recognition unit 11 may be configured to acquire the information on the width ΔW of the target lane 51 and use the width ΔW of the target lane 51 for recognition of the first relative distance L1. The feature recognition unit 11 acquires the information about the width ΔW of the target lane 51, for example, by referring to the storage device 3 or by communication such as road-vehicle communication. When the target road 5 has a plurality of lanes 50 having the same traveling direction M on the widthwise first side W1 with respect to the separated feature 6 (that is, when the target road 5 has a plurality of target lanes 51). The width ΔW of the target lane 51 is the sum of the lane widths of the plurality of target lanes 51. An end portion (lane end 51a) of the width-direction first side W1 in the target lane 51 (the target lane 51 of the width-direction first side W1 when the target road 5 has a plurality of target lanes 51), and the target vehicle The relative distance along the road width direction W with respect to the reference position in 2 is set as the second relative distance L2 (see FIG. 2), for example, the feature recognition unit 11 determines the width ΔW of the target lane 51 and the second relative distance L2. The first relative distance L1 can be recognized based on

As shown in FIG. 2, when the reference position of the target vehicle 2 is located on the second side W2 in the width direction with respect to the lane end 51a, the second relative distance L2 is subtracted from the width ΔW of the target lane 51 to obtain the first value. The relative distance L1 can be derived, and although not shown, when the reference position of the target vehicle 2 is located on the width-direction first side W1 with respect to the lane end 51a, the second distance ΔW of the target lane 51 The relative distance L2 can be added to derive the first relative distance L1. The second relative distance L2 is, for example, an image of a lane end 51a included in an image captured by the camera 24 or another camera or a feature near the lane end 51a (for example, a road marking provided on the lane end 51a or in the vicinity thereof). It can be acquired based on the recognition result.

The determination unit 12 determines whether or not the target vehicle 2 can perform the forward running approach in which the target vehicle lane 51 enters the target lane 51 and travels in the forward running direction only by traveling forward. This is a functional unit that determines that the target vehicle 2 may reverse in the target lane 51 when it is determined based on the distance L1 that the target vehicle 2 cannot make a forward approach. When the target road 5 has a plurality of target lanes 51, the target vehicle 2 should enter the target lane 51 at least on the second widthwise second side W2 and travel in the forward direction only when traveling forward. When the target vehicle 2 can make a forward approach, it is determined that the target vehicle 2 can make a forward approach. In the present embodiment, the process executed by the determination unit 12 corresponds to the “determination step”, and the function realized by executing the process corresponds to the “determination function”.

In the present embodiment, the determination unit 12 determines the target when the target vehicle 2 turns at the minimum turning radius R of the target vehicle 2 (turns to the second side H2) based on the relative angle θ1 and the first relative distance L1. When the movement trajectory T of the vehicle 2 (hereinafter, simply referred to as the movement trajectory T of the target vehicle 2) overlaps with the separated feature 6, it is determined that the target vehicle 2 cannot make a forward approach, and the target vehicle 2 When the movement trajectory T of does not overlap the separated feature 6, it is determined that the target vehicle 2 can make a forward approach. The movement trajectory T of the target vehicle 2 is the minimum of the target vehicle 2 from the position and the direction (the position and the direction of the target vehicle 2 shown by the broken line in FIG. 2) at the time when the relative angle θ1 and the first relative distance L1 are recognized. It is a movement locus of the target vehicle 2 when turning with a turning radius R.

In FIG. 3, it is assumed that the movement trajectory T of the target vehicle 2 does not overlap the separated feature 6, and in this case, the determination unit 12 determines that the target vehicle 2 can make a forward approach. FIG. 3 shows an example of a time-series change in the position of the target vehicle 2 when the target vehicle 2 enters the target lane 51 in a forward direction. Time t1, time t2, time t3, and time are shown in order from the earliest time. It is t4. Even if the direction of the target vehicle 2 when entering the target road 5 has a component opposite to the traveling direction M of the target lane 51, the target vehicle 2 makes a forward approach as shown in FIG. You may be able to. In the present embodiment, the determination unit 12 is configured not to determine that the target vehicle 2 may reverse in the target lane 51 when the target vehicle 2 can make a forward approach. .. As a result, it is possible to reduce the possibility of an erroneous determination that the driver may reverse when the driver is performing a normal driving operation. It should be noted that “not determining that the target vehicle 2 may reverse the target lane 51” means that the target vehicle 2 may reverse the target lane 51 when the target vehicle 2 enters the target road 5. It means that the target vehicle 2 does not determine that there is a possibility that the target vehicle 2 may run backward in the target lane 51 after the target vehicle 2 enters the target road 5.

On the other hand, in FIG. 4, it is assumed that the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6 as shown in FIG. 2, and in this case, the determination unit 12 causes the target vehicle 2 to perform a forward approach. It is determined that it is not possible. FIG. 4 shows an example of a time-series change in the position of the target vehicle 2 when the target vehicle 2 travels backward in the target lane 51, and is time t5, time t6, and time t7 in order from the earliest time. There is. When the target vehicle 2 cannot make a forward approach, the target vehicle 2 that has entered the target road 5 is likely to reverse the target lane 51, as in the example shown in FIG. As described above, the determination unit 12 determines that the target vehicle 2 may run backward in the target lane 51 when it is determined that the target vehicle 2 cannot make a forward approach. As a result, it is possible to reduce the possibility of an erroneous determination that does not determine that the target vehicle 2 may reverse in a situation in which the target vehicle 2 is likely to reverse in the target lane 51. ..

Whether or not the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6 will be described below with reference to FIG. 2, in which the relative angle θ1 when the target vehicle 2 enters the target road 5 and the first It can be determined based on the relative distance L1. As is apparent from FIG. 2, the road width direction of the turning center C with respect to the target road 5 when the target vehicle 2 turns to the second side H2 with the minimum turning radius R (minimum turning radius) (here, left turn). The position of W is determined according to the relative angle θ1 and the first relative distance L1. As is apparent from FIG. 3 and FIG. 4 in addition to FIG. 2, the turning center C moves to the width direction second side W2 as the relative angle θ1 increases, and as the first relative distance L1 decreases. It moves to the width direction second side W2. Then, as shown in FIG. 2, when the posture angle θ3 determined according to the position of the turning center C and the size of the minimum turning radius R is used, a portion that draws a locus of the minimum turning radius R in the target vehicle 2 (in FIG. 2, The maximum moving distance Z of the right front wheel of the target vehicle 2) along the road width direction W from the lane end 51a to the width direction second side W2 is calculated by using the minimum turning radius R and the posture angle θ3 as {R(1+sin θ3 )}. When the maximum movement distance Z is larger than the width ΔW of the target lane 51, the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6.

When the target vehicle 2 turns at the minimum turning radius R, the radius of the outer circumference of the movement trajectory T of the target vehicle 2 (the radius of the dashed-dotted circle in FIG. 2) is different from the example shown in FIG. When the maximum travel distance Z is equal to or smaller than the width ΔW of the target lane 51, the travel trajectory T of the target vehicle 2 does not overlap the separated feature 6. On the other hand, in FIG. 2, the minimum vehicle body turning radius (outer vehicle body minimum turning radius) of the target vehicle 2 is larger than the minimum turning radius R, and the vehicle body minimum turning radius is the outer radius of the movement trajectory T of the target vehicle 2. Is assumed to match. In this case, when the value obtained by adding the difference between the minimum turning radius of the vehicle body and the minimum turning radius R to the maximum travel distance Z (hereinafter referred to as the corrected maximum travel distance) is less than or equal to the width ΔW of the target lane 51, The movement trajectory T of 2 does not overlap with the separated feature 6. Even if the maximum travel distance Z is less than or equal to the width ΔW of the target lane 51, if the corrected maximum travel distance is greater than the width ΔW of the target lane 51, the travel trajectory T of the target vehicle 2 is a separated feature. It overlaps with 6. That is, in the present embodiment, when the corrected maximum movement distance is larger than the width ΔW of the target lane 51, it is determined that the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6, and the corrected maximum movement distance of the target lane 51 is determined. When the width is equal to or smaller than the width ΔW, it is determined that the movement trajectory T of the target vehicle 2 does not overlap the separated feature 6.

The corrected maximum movement distance is not calculated by adding the difference between the minimum turning radius R and the minimum turning radius R to the maximum moving distance Z, but by using the minimum turning radius R instead of the minimum turning radius R in {R(1+sin θ3)}. You may derive using it. In this case, the posture angle θ3 is an angle when the size of the minimum turning radius R in FIG. 2 is changed to the size of the minimum turning radius of the vehicle body. Further, in the case where the difference between the minimum turning radius of the vehicle body and the minimum turning radius R is small, and the maximum movement distance Z is larger than the width ΔW of the target lane 51, the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6. If the maximum travel distance Z is less than or equal to the width ΔW of the target lane 51, it may be determined that the travel trajectory T of the target vehicle 2 does not overlap the separated feature 6. The determination unit 12 uses, for example, the vehicle information of the target vehicle 2 (information about the minimum turning radius R and the minimum turning radius of the vehicle body) that is used to determine whether the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6. It is acquired by referring to the storage device 3. Further, the determination unit 12 acquires information on the width ΔW of the target lane 51 used for determining whether or not the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6, for example, by referring to the storage device 3, Alternatively, it is acquired by communication such as road-to-vehicle communication.

In this embodiment, the determination unit 12 determines whether or not the target vehicle 2 can make a forward approach when the relative angle θ1 is larger than 90 degrees. That is, the determination unit 12 determines whether or not the movement trajectory T of the target vehicle 2 overlaps the separated feature 6 when the relative angle θ1 is larger than 90 degrees. Therefore, in the present embodiment, when the relative angle θ1 is 90 degrees or less, the feature recognition unit 11 does not need to recognize the first relative distance L1. Therefore, for example, when the relative angle θ1 is greater than 90 degrees, the feature recognition unit 11 recognizes the first relative distance L1, and when the relative angle θ1 is 90 degrees or less, the feature recognition unit 11 determines the first relative distance L1. The configuration may be such that the one relative distance L1 is not recognized.

In the present embodiment, the determination unit 12 determines that the relative angle θ1 is 90 when the extending direction E of the separated feature 6 in the captured image 7 is inclined toward the upper side V toward the first side H1. It is judged to be greater than the degree. Therefore, in the present embodiment, when the extending direction E of the separated feature 6 in the captured image 7 is not inclined toward the upper side V toward the first side H1, it is necessary to derive the relative angle θ1. There is no. On the other hand, when the extending direction E of the separated features 6 in the captured image 7 is inclined toward the upper side V toward the first side H1, the relative angle θ1 is determined for the determination by the determination unit 12. Need to be derived. Therefore, for example, when the extending direction E of the separated feature 6 in the captured image 7 is inclined toward the upper side V toward the first side H1, the feature recognizing unit 11 derives the relative angle θ1. When the extending direction E of the separated feature 6 in the captured image 7 is not inclined toward the upper side V toward the first side H1, the feature recognizing unit 11 derives the relative angle θ1. It can be configured not to perform.

The warning processing unit 14 is a functional unit that issues a warning when the determination unit 12 determines that the target vehicle 2 may run backward in the target lane 51. The warning processing unit 14 issues a warning using the warning device 4 when the determination unit 12 determines that the target vehicle 2 may reversely travel in the target lane 51. The warning processing unit 14 executes a process (warning process step) of issuing a warning when it is determined that the target vehicle 2 may reversely travel in the target lane 51. A function (warning processing function) of issuing a warning when it is determined that the vehicle may reversely run in the target lane 51 is realized.

The warning processing unit 14 uses, for example, a display as the warning device 4, displays warning character information and a blinking pattern of light on the display, and warns an occupant (particularly, a driver) of the target vehicle 2. To do. Further, the warning processing unit 14 uses, for example, a speaker as the warning device 4, and issues a warning message or warning sound for warning from the speaker to warn the occupant of the target vehicle 2.

Hereinafter, with reference to FIG. 5, a processing procedure of the reverse running determination process executed in the reverse running determination system 1 of the present embodiment will be described. Each step described below is executed by an arithmetic processing unit (computer) included in the reverse running determination system 1.

As shown in FIG. 5, the control unit 10 determines whether or not the target vehicle 2 is about to enter the target road 5 from the outside in the road width direction W (outside the width direction first side W1). #01). When it is determined that the target vehicle 2 is approaching the target road 5 from the outside in the road width direction W (step #01: Yes), the feature recognition processing by the feature recognition unit 11 is performed. (Step #02). In step #02, it is possible to determine to a small extent whether or not the extending direction E of the separated features 6 in the captured image 7 is inclined toward the upper side V toward the first side H1. Recognition processing of the separated features 6 is performed.

The determination unit 12 determines whether the relative angle θ1 is greater than 90 degrees (step #03). In the present embodiment, the determination unit 12 determines that the relative angle θ1 is 90 when the extending direction E of the separated feature 6 in the captured image 7 is inclined toward the upper side V toward the first side H1. It is judged to be greater than the degree. When it is determined that the relative angle θ1 is larger than 90 degrees (step #03: Yes), the determination unit 12 determines whether the target object is based on the recognition result of the relative angle θ1 and the first relative distance L1 by the feature recognition unit 11. It is determined whether or not the movement trajectory T of the vehicle 2 overlaps with the separated feature 6 (step #04). On the other hand, when it is determined that the relative angle θ1 is 90 degrees or less (step #03: No), the process ends. The recognition of the relative angle θ1 and the first relative distance L1 by the feature recognition unit 11 may be performed in step #02 instead of between step #03 and step #04.

When it is determined that the movement trajectory T of the target vehicle 2 overlaps with the separated feature 6 (step #04: Yes), the determination unit 12 determines that the target vehicle 2 cannot make a forward approach ( In step #05), the warning processing unit 14 performs warning processing using the warning device 4 (step #06), and the processing ends. On the other hand, when it is determined that the movement trajectory T of the target vehicle 2 does not overlap the separated feature 6 (step #04: No), the process ends.

[Other Embodiments]
Next, another embodiment of the reverse running determination system will be described.

(1) In the above-described embodiment, when the determination unit 12 tilts the extending direction E of the separated features 6 in the captured image 7 toward the upper side V toward the first side H1, it is relatively determined. The configuration in which the angle θ1 is determined to be larger than 90 degrees has been described as an example. However, the configuration is not limited to such a configuration, and the determination unit 12 determines whether the relative angle θ1 is greater than 90 degrees based on the recognition result of the relative angle θ1 by the feature recognition unit 11. You can also

(2) In the above embodiment, the determination unit 12 has been described as an example of the configuration that determines whether or not the target vehicle 2 can make a forward approach when the relative angle θ1 is larger than 90 degrees. However, without being limited to such a configuration, when the determination unit 12 determines that the set angle is an angle greater than or equal to 0 degree and less than 180 degrees and other than 90 degrees, the relative angle θ1 is larger than the set angle. Alternatively, it may be configured to determine whether or not the target vehicle 2 can make a forward approach. Further, the determination unit 12 may be configured to determine whether or not the target vehicle 2 can make a forward approach regardless of the magnitude of the relative angle θ1.

(3) In the above embodiment, the feature recognition unit 11 acquires the captured image 7 of the front of the target vehicle 2 by the camera 24, and performs the image recognition process of the separated features 6 included in the captured image 7, The configuration in which the relative angle θ1 is recognized based on the tilt angle θ2 of the separated feature 6 in the extending direction E in the captured image 7 has been described as an example. However, without being limited to such a configuration, for example, the information acquisition unit 20 includes a radar (laser radar, millimeter wave radar, or the like) that detects an object at least in front of the target vehicle 2, and the feature recognition unit 11 However, the relative angle θ1 may be recognized based on the detection result of the radar. Similarly, unlike the above embodiment, the feature recognition unit 11 may be configured to recognize the first relative distance L1 based on the detection result of the radar. In such a case, as the radar, it is preferable to use a radar (scan type radar) capable of detecting the distance to the object and the direction in which the object exists.

(4) In the above-described embodiment, the reverse running determination system 1 (control unit 10) includes the warning processing unit 14, and the determination unit 12 determines that the target vehicle 2 may run backward in the target lane 51. In this case, the configuration in which the warning processing unit 14 issues a warning has been described as an example. However, without being limited to such a configuration, in the case where the determination unit 12 determines that the target vehicle 2 may reversely run in the target lane 51, instead of or in addition to the warning process, Alternatively, the behavior of the target vehicle 2 may be controlled (for example, control for braking the target vehicle 2 is performed).

(5) The allocation of each functional unit of the reverse running determination system 1 (control unit 10) shown in the above embodiment is merely an example, and a plurality of functional units may be combined or one functional unit may be further divided. It is also possible.

(6) Note that the configuration disclosed in each of the above-described embodiments may be applied in combination with the configuration disclosed in another embodiment as long as no contradiction occurs (a difference between the embodiments described as other embodiments. (Including combinations) is also possible. Regarding other configurations, the embodiments disclosed in the present specification are merely examples in all respects. Therefore, various modifications can be appropriately made without departing from the spirit of the present disclosure.

[Outline of the above embodiment]
The outline of the reverse running determination system described above will be described below.

The reverse running determination system (1) sets a road having a separated feature (6), which is a feature that separates lanes (50) having different traveling directions (M), as a target road (5), and the target road (5) ), the target vehicle (2) has the lane (50) on the first side (W1) in the width direction, which is one side of the road width direction (W) with respect to the separated feature (6), as the target lane (51). When approaching the target road (5) from the outside of the first side (W1) in the width direction, the vehicle body front-back direction (D) of the target vehicle (2) and the separated feature (6) A feature recognition unit (11) that recognizes a relative angle (θ1) with respect to the extending direction (E) and a relative distance (L1) between the target vehicle (2) and the separated feature (6), and a forward movement. Whether or not the target vehicle (2) can enter the forward lane in which the vehicle enters the target lane (51) and travels in the forward direction only by traveling is determined by the relative angle (θ1) and When it is determined based on the relative distance (L1) that the target vehicle (2) cannot perform the forward approach, the target vehicle (2) reverses the target lane (51). A determination unit (12) for determining that there is a possibility of running.

Depending on the direction and position of the target vehicle (2) when entering the target road (5), the travel line that the target vehicle (2) can take after entering the target road (5) is determined. ) Can perform forward approach based on the direction and position of the target vehicle (2) when entering the target road (5). According to the above configuration, the relative angle (θ1) and the relative distance (L1) representing the direction and position of the target vehicle (2) when entering the target road (5) are provided on the target road (5). It is possible to acquire the separated feature (6) as a reference and determine whether or not the target vehicle (2) can perform forward approach based on the relative angle (θ1) and the relative distance (L1). it can.

Then, in the above configuration, when it is determined that the target vehicle (2) cannot make a forward approach, it is determined that the target vehicle (2) may run backward in the target lane (51). It When the target vehicle (2) cannot make a forward approach, the target vehicle (2) entering the target road (5) is likely to reverse the target lane (51). According to the above configuration, in a situation in which the target vehicle (2) is likely to reversely run in the target lane (51), it is possible to reduce the possibility of an erroneous determination that the reverse running is not determined. be able to. On the other hand, in the above configuration, when the target vehicle (2) can make a forward approach, it is not determined that the target vehicle (2) may run backward in the target lane (51). As a result, it is possible to reduce the possibility of erroneous determination that the driver may reverse when the driver is performing a normal driving operation.

As described above, according to the above configuration, in the case of determining the possibility that the target vehicle (2) trying to enter the target road (5) from the outside of the first side (W1) in the width direction may reversely travel. In, it is possible to reduce the possibility of erroneous determination.

Here, the feature recognition unit (11) acquires a captured image (7) in front of the target vehicle (2) by the camera (24), and the separated feature included in the captured image (7). It is preferable to perform the image recognition process of (6) and recognize the relative angle (θ1) based on the inclination angle (θ2) of the separated feature (6) in the extending direction (E) of the captured image (7). is there.

According to this configuration, the target vehicle (2) is subjected to the image recognition processing of the captured image (7) based on the correspondence relationship between each position in the captured image (7) by the camera (24) and the actual position on the road surface. The relative angle (θ1) between the vehicle front-rear direction (D) and the extension direction (E) of the separated feature (6) can be recognized.

Further, a vector (A1) directed forward of the target vehicle (2) along the vehicle front-rear direction (D) and the target lane (51) along the extending direction (E) of the separated feature (6). ), the angle formed by the vector (A2) directed to the traveling direction (M) is defined as the relative angle (θ1), and when the relative angle (θ1) is greater than 90 degrees, the determination unit (12) determines It is preferable to determine whether or not the target vehicle (2) can perform the forward approach.

When the relative angle (θ1) is larger than 90 degrees, the direction of the target vehicle (2) when entering the target road (5) is the component opposite to the traveling direction (M) of the target lane (51). Therefore, the possibility that the target vehicle (2) will reverse the target lane (51) after entering the target road (5) is significantly higher than that when the relative angle (θ1) is 90 degrees or less. It is considered to be. According to the above configuration, the scene in which the target vehicle (2) determines whether or not the vehicle can enter the forward running is limited to the case where the relative angle (θ1) is larger than 90 degrees, and thus the target vehicle It is possible to reduce the processing load while suppressing the possibility of erroneous determination regarding the possibility that (2) may run backward.

As described above, the determination unit (12) determines whether the target vehicle (2) can perform the forward approach when the relative angle (θ1) is greater than 90 degrees. In the above, the feature recognition unit (11) acquires a captured image (7) in front of the target vehicle (2) by the camera (24), and the separated feature (7) included in the captured image (7). The image recognition processing of 6) is performed, and when the left side traffic is defined by the traffic rules, the right side is defined as the first side (H1) when the right side traffic is defined by the traffic rules. The determination unit (12) inclines so that the extending direction (E) of the separated feature (6) in the captured image (7) is directed toward the upper side (V) as it extends toward the first side (H1). In this case, it is preferable to determine that the relative angle (θ1) is larger than 90 degrees.

According to this configuration, when the relative angle (θ1) is larger than 90 degrees, the extending direction (E) of the separated feature (6) in the captured image (7) is increased toward the first side (H1). Paying attention to the inclination toward (V), it is possible to determine whether or not the relative angle (θ1) is larger than 90 degrees without deriving the value of the relative angle (θ1). Therefore, the processing load can be further reduced.

In the reverse running determination system (1) having each of the above configurations, the determination unit (12) determines that the target vehicle (2) is the target vehicle (2) based on the relative angle (θ1) and the relative distance (L1). If the movement trajectory (T) of the target vehicle (2) when turning with the minimum turning radius (R) of 2) overlaps with the separated feature (6), the forward approach cannot be performed. It is preferable to determine that the forward running can be performed when the movement trajectory (T) does not overlap the separated feature (6).

According to this configuration, it is possible to appropriately determine whether or not the target vehicle (2) can perform forward approach based on the movement trajectory (T) when the target vehicle (2) makes the smallest turn. ..

Further, the feature recognition unit (11) acquires information on the width (ΔW) of the target lane (51) and recognizes the width (ΔW) of the target lane (51) as the relative distance (L1). It is suitable to be used for.

According to this configuration, the accuracy of recognizing the relative distance (L1) can be increased more easily than in the case where the width (ΔW) of the target lane (51) is not used for recognizing the relative distance (L1).

The reverse running determination system according to the present disclosure may have at least one of the effects described above.

The various technical features of the reverse running determination system described above are also applicable to a reverse run determining method and a reverse run determining program, and such a method and program, and a storage medium storing such a program. (Eg, optical disks, flash memory, etc.) are also disclosed herein. For example, the reverse run determination method can include steps with the features of the reverse run determination system described above. Further, the reverse running determination program can cause a computer to realize the function having the features of the reverse running determination system described above. As a matter of course, these reverse run determination method and reverse run determination program can also achieve the operational effects of the above-described reverse run determination system. Furthermore, it is also possible to incorporate various additional features exemplified as a preferable aspect of the reverse run determination system into these reverse run determination methods and reverse run determination programs, and the method and the program have respective additional features. Corresponding operational effects can also be achieved.

1: Reverse running determination system 2: Target vehicle 5: Target road 6: Separated features 7: Photographed image 11: Feature recognition unit 12: Determination unit 24: Camera 50: Lane 51: Target lane A1: First vector (vehicle body Vector that goes forward of the target vehicle along the front-back direction)
A2: Second vector (vector extending in the traveling direction of the target lane along the extending direction of the separated features)
D: vehicle front-rear direction E: extension direction H1: first side L1: first relative distance (relative distance)
M: traveling direction R: minimum turning radius T: movement trajectory V: upper side W: road width direction W1: first side in width direction ΔW: width θ1: relative angle θ2: inclination angle

Claims (8)

A road having a separated feature that is a feature that separates lanes having different traveling directions is a target road, and a lane on the first side in the width direction that is one side of the road width direction with respect to the separated feature on the target road. As the target lane,
When the target vehicle is approaching the target road from the outside on the first side in the width direction, the relative angle between the vehicle body front-rear direction of the target vehicle and the extending direction of the separated features, and the target A feature recognition unit that recognizes a relative distance between the vehicle and the separated feature,
Based on the relative angle and the relative distance, it is determined whether or not the target vehicle can perform a forward approach that enters the target lane and travels in the forward direction only by traveling forward. A reverse running determination system that determines that the target vehicle may reverse in the target lane when it is determined that the target vehicle cannot make the forward approach. .. The feature recognition unit acquires a captured image of the front of the target vehicle by a camera, performs image recognition processing of the separated feature included in the captured image, and extends the separated feature in the captured image. The reverse running determination system according to claim 1, wherein the relative angle is recognized based on a tilt angle of the direction. As a relative angle, an angle formed by a vector heading forward of the target vehicle along the vehicle body front-rear direction and a vector heading in the traveling direction of the target lane along the extending direction of the separated features,
The reverse running determination system according to claim 1, wherein the determination unit determines whether the target vehicle can perform the forward running approach when the relative angle is larger than 90 degrees. The feature recognition unit acquires a captured image of the front of the target vehicle by a camera, performs image recognition processing of the separated features included in the captured image,
If the left side traffic is defined by the traffic rules, the right side is defined as the first side if the right side traffic is defined by the traffic rules,
The determination unit determines that the relative angle is larger than 90 degrees when the extending direction of the separated feature in the captured image is inclined so as to be directed upward as it extends toward the first side, The reverse running determination system according to claim 3. The determination unit, based on the relative angle and the relative distance, when the movement trajectory of the target vehicle when the target vehicle turns at the minimum turning radius of the target vehicle overlaps with the separated feature, 5. It is determined that the forward approach cannot be performed, and it is determined that the forward approach can be performed when the movement locus does not overlap with the separated feature. Reverse run determination system described. The reverse running determination system according to any one of claims 1 to 5, wherein the feature recognition unit acquires information on the width of the target lane and uses the width of the target lane for recognition of the relative distance. .. A road having a separated feature that is a feature that separates lanes having different traveling directions is a target road, and a lane on the first side in the width direction that is one side of the road width direction with respect to the separated feature on the target road. As the target lane,
When the target vehicle is approaching the target road from the outside on the first side in the width direction, the relative angle between the vehicle body front-rear direction of the target vehicle and the extending direction of the separated features, and the target A feature recognition step of recognizing a relative distance between the vehicle and the separated feature;
Based on the relative angle and the relative distance, it is determined whether or not the target vehicle can perform a forward approach that enters the target lane and travels in the forward direction only by traveling forward. A reverse running determining method, which includes a step of determining that the target vehicle may reversely run in the target lane when it is determined that the target vehicle cannot perform the forward approach. .. A road having a separated feature that is a feature that separates lanes having different traveling directions is a target road, and a lane on the first side in the width direction that is one side of the road width direction with respect to the separated feature on the target road. As the target lane,
When the target vehicle is approaching the target road from the outside on the first side in the width direction, the relative angle between the vehicle body front-rear direction of the target vehicle and the extending direction of the separated features, and the target A feature recognition function that recognizes a relative distance between a vehicle and the separated feature,
Based on the relative angle and the relative distance, it is determined whether or not the target vehicle can perform a forward approach that enters the target lane and travels in the forward direction only by traveling forward. A determination function for determining that the target vehicle may reversely run in the target lane when it is determined that the target vehicle cannot make the forward approach. , Reverse run determination program.

Images