Target vehicle recognition device and processing method for target vehicle recognition device

The target vehicle recognition device uses a front camera and radar sensor to accurately identify stopped vehicles for steering control by determining orientation and boundary line deviation, preventing unnecessary steering controls and ensuring safe navigation.

JP7589664B2Active Publication Date: 2024-11-26TOYOTA JIDOSHA KK
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
JP2021158260
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-09-28
Publication Date
2024-11-26
Estimated Expiration
2041-09-28

AI Technical Summary

Technical Problem

Existing vehicle recognition systems inaccurately identify moving vehicles as stopped vehicles when they straddle lane boundaries, leading to unnecessary steering controls, especially when vehicles are traveling around curves, due to decreased speed detection accuracy and image distortion.

Method used

A target vehicle recognition device using a combination of a front camera and radar sensor to detect stopped vehicles, determine their orientation and boundary line deviation states, and accurately identify vehicles for steering control by analyzing the presence of vehicle edges in the captured image.

Benefits of technology

Prevents erroneous recognition of moving vehicles as stopped vehicles, allowing for appropriate steering control and safe navigation around curves by accurately distinguishing between stopped and moving vehicles based on edge detection and orientation analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007589664000001
    Figure 0007589664000001
  • Figure 0007589664000002
    Figure 0007589664000002
  • Figure 0007589664000003
    Figure 0007589664000003
Patent Text Reader

Abstract

To properly recognize an object vehicle being an avoidance object of steering control from a photographed image.SOLUTION: An object vehicle recognition device for recognizing an object vehicle being an avoidance object of steering control of an own vehicle comprises: a stop vehicle detection unit which detects a stop vehicle stopping in front of the own vehicle; an orientation determination unit which determines whether the stop vehicle is in the front-facing state in which the front face of the stop vehicle is oriented to the own vehicle or in the rear-facing state in which the rear face of the stop vehicle is oriented to the own vehicle; a boundary line deviation state determination unit which determines whether the stop vehicle is in the right boundary line deviation state in which the stop vehicle is across the boundary line on the right side of a travel lane of the own vehicle or the left boundary line deviation state in which the stop vehicle is across the boundary line on the left side of the travel lane or the stop vehicle is not in any of the right boundary line deviation state and the left boundary line deviation state; and an object vehicle recognition unit which recognizes the object vehicle of steering control on the basis of the photographed image taken by a front camera of the own vehicle, the determination result of the orientation determination unit and the determination result of the boundary line deviation state determination unit.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a target vehicle recognition device and a processing method for the target vehicle recognition device. [Background technology]

[0002] Conventionally, JP2019-524525A is known as a technical document related to a target vehicle recognition device for recognizing a target vehicle to be avoided by steering control of the vehicle. This publication discloses that, in consideration of the possibility that a pedestrian may jump out from the shadow of a parked vehicle parked protruding into the driving lane of the vehicle, the vehicle is steered to pass the side of the parked vehicle at a distance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special table 2019-524525 publication Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, in order to recognize a target vehicle to be avoided by steering control, it has been considered to use an image captured by a front camera of the vehicle. The above-mentioned publication does not disclose a method for accurately recognizing a target vehicle to be avoided by steering control (a stopped vehicle that is stopped protruding from the road shoulder into the driving lane of the vehicle). In particular, when a preceding vehicle on the driving lane of the vehicle enters a curve, the preceding vehicle may appear to be located straddling the boundary line of the driving lane in an image captured from the vehicle. In addition, the greater the distance from the preceding vehicle, the lower the detection accuracy of the speed of the preceding vehicle, and there is a possibility that the preceding vehicle will be mistakenly recognized as a stopped vehicle. In such a case, if the preceding vehicle is mistakenly recognized as a target vehicle for steering control, the vehicle will perform unnecessary steering control, so a method for accurately recognizing a target vehicle for steering control is required. [Means for solving the problem]

[0005] One aspect of the present invention is a target vehicle recognition device for recognizing a target vehicle to be avoided by steering control of a host vehicle, the target vehicle recognition device including a stopped vehicle detection unit that detects a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle, an orientation determination unit that determines whether the stopped vehicle is in a forward-facing state with its front facing the host vehicle or in a rear-facing state with its back facing the host vehicle based on the image captured by the front camera of the host vehicle, and a right-side boundary determination unit that determines whether the stopped vehicle is straddling a right boundary line of the driving lane of the host vehicle based on the image captured by the front camera of the host vehicle. a boundary line departure state determination unit that determines whether the stopped vehicle is in a boundary line departure state, whether the stopped vehicle is in a left boundary line departure state straddling the boundary line on the left side of the travel lane, or whether the stopped vehicle is in neither a right boundary line departure state nor a left boundary line departure state, and a target vehicle recognition unit that recognizes a target vehicle for steering control based on an image captured by a front camera of the host vehicle, a determination result from the orientation determination unit, and a determination result from the boundary line departure state determination unit, and the target vehicle recognition unit recognizes the stopped vehicle as a target vehicle for steering control when the boundary line departure state determination unit determines that the stopped vehicle is neither a right boundary line departure state nor a left boundary line departure state. When the direction determination unit determines that the stopped vehicle is in a backward facing state and the boundary line departure state determination unit determines that the stopped vehicle is in a left boundary line departure state, and when the front left end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and when the front left end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control, and when the direction determination unit determines that the stopped vehicle is in a backward facing state and the boundary line departure state determination unit determines that the stopped vehicle is in a right boundary line departure state, and when the front right end of the stopped vehicle is not included in the captured image , the stopped vehicle is recognized as a target vehicle for steering control, and when the front right edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control and the stopped vehicle is determined to be in a forward-facing state by the orientation determination unit, and when the boundary line departure state determination unit determines that the stopped vehicle is in a left boundary line departure state, when the rear right edge of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and when the rear right edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control and the stopped vehicle is determined to be in a forward-facing state by the orientation determination unit,In addition, when the boundary line departure state determination unit determines that the stopped vehicle is in a right boundary line departure state, if the rear left edge of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a vehicle to be subjected to steering control, and if the rear left edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a vehicle to be subjected to steering control.

[0006] According to a target vehicle recognition device of one aspect of the present invention, when a vehicle preceding or an oncoming vehicle is traveling around a curve, even if the vehicle is mistakenly determined to be a stopped vehicle and the preceding or oncoming vehicle is determined to be in a state of deviating from the left boundary line or the right boundary line in the image captured by the vehicle's front camera, by determining whether the captured image includes any of the edges of the four corners of the preceding or oncoming vehicle traveling around the curve that are not visible in the captured image, the device can avoid mistakenly recognizing the preceding or oncoming vehicle traveling around the curve as a vehicle subject to steering control and can properly recognize target vehicles that are to be avoided through steering control from the captured image.

[0007] Another aspect of the present invention is a target vehicle recognition device for recognizing a target vehicle that is to be avoided by steering control of the host vehicle, the target vehicle recognition device including a stopped vehicle detection unit that detects a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle, an orientation determination unit that determines whether the stopped vehicle is in a forward-facing state with its front facing the host vehicle or in a rear-facing state with its back facing the host vehicle based on the image captured by the front camera of the host vehicle, and a right-facing state that determines whether the stopped vehicle is straddling the right boundary line of the driving lane of the host vehicle based on the image captured by the front camera of the host vehicle. The vehicle is provided with a boundary line departure state determination unit that determines whether the stopped vehicle is in a boundary line departure state, whether the stopped vehicle is in a left boundary line departure state straddling the boundary line on the left side of the driving lane, or whether the stopped vehicle is in neither a right boundary line departure state nor a left boundary line departure state, and a target vehicle recognition unit that recognizes a target vehicle for steering control based on an image captured by a front camera of the vehicle, a determination result from the orientation determination unit, and a determination result from the boundary line departure state determination unit, and the target vehicle recognition unit recognizes the stopped vehicle as a target vehicle for steering control when the boundary line departure state determination unit determines that the stopped vehicle is neither a right boundary line departure state nor a left boundary line departure state. When the direction determining unit determines that the stopped vehicle is in a backward facing state and the boundary line departure state determining unit determines that the stopped vehicle is in a left boundary line departure state, if the front right end of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control, and when the front right end of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and when the direction determining unit determines that the stopped vehicle is in a backward facing state and the boundary line departure state determining unit determines that the stopped vehicle is in a right boundary line departure state, the front left end of the stopped vehicle is not included in the captured image. when the stopped vehicle is not recognized as a target vehicle for steering control, and when the front left edge of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and the stopped vehicle is determined to be in a forward-facing state by the orientation determination unit, and when the boundary line departure state determination unit determines that the stopped vehicle is in a left boundary line departure state, when the rear left edge of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control, and when the rear left edge of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and the stopped vehicle is determined to be in a forward-facing state by the orientation determination unit,In addition, when the boundary line departure state determination unit determines that the stopped vehicle is in a right boundary line departure state, if the rear right edge of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a vehicle to be subjected to steering control, and if the rear right edge of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a vehicle to be subjected to steering control.

[0008] According to a target vehicle recognition device of another aspect of the present invention, when a vehicle preceding or an oncoming vehicle is traveling around a curve, even if it is mistakenly determined to be a stopped vehicle and the preceding or oncoming vehicle is determined to be in a state of deviating from the left boundary line or the right boundary line in the image captured by the front camera of the vehicle, by determining whether the captured image includes any of the edges of the four corners of the preceding or oncoming vehicle traveling around the curve that are not visible in the captured image, it is possible to avoid mistakenly recognizing the preceding or oncoming vehicle traveling around the curve as a vehicle subject to steering control and to properly recognize target vehicles that are to be avoided through steering control from the captured image.

[0009] Yet another aspect of the present invention is a processing method of a target vehicle recognition device for recognizing a target vehicle that is to be avoided by steering control of the host vehicle, the method including: a stopped vehicle detection step of detecting a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle; an orientation determination step of determining, based on the image captured by the front camera of the host vehicle, whether the stopped vehicle is in a forward-facing state with its front facing the host vehicle or in a backward-facing state with its back facing the host vehicle; a boundary line departure state determination step of determining, based on the image captured by the front camera of the host vehicle, whether the stopped vehicle is in a right boundary line departure state crossing the boundary line on the right side of the traveling lane of the host vehicle, whether the stopped vehicle is in a left boundary line departure state crossing the boundary line on the left side of the traveling lane, or whether the stopped vehicle is in neither a right boundary line departure state nor a left boundary line departure state; and a target vehicle recognition step of recognizing a target vehicle for steering control based on the image captured by the front camera of the host vehicle, the determination result of the orientation determination step, and the determination result of the boundary line departure state determination step. In the target vehicle recognition step, When it is determined that the stopped vehicle is not in either of the line departure states, the stopped vehicle is not recognized as a target vehicle for steering control, and when it is determined that the stopped vehicle is in a backward facing state in the direction determination step and when it is determined that the stopped vehicle is in a left boundary line departure state in the boundary line departure state determination step, if the front left end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and when the front left end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control, and it is determined that the stopped vehicle is in a backward facing state in the direction determination step and when it is in a boundary line departure state determination step, When the state determination step determines that the stopped vehicle is in a right boundary line departure state, and if the front right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a vehicle to be subjected to steering control; when the front right end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a vehicle to be subjected to steering control; when the direction determination step determines that the stopped vehicle is in a forward-facing state and the boundary line departure state determination step determines that the stopped vehicle is in a left boundary line departure state, and if the rear right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a vehicle to be subjected to steering control;When the rear right edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control, and when the stopped vehicle is determined to be in a forward-facing state in the direction determination step and the stopped vehicle is determined to be in a right boundary line departure state in the boundary line departure state determination step, if the rear left edge of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for steering control, and when the rear left edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for steering control.

[0010] According to a processing method of a target vehicle recognition device relating to yet another aspect of the present invention, when a vehicle preceding or an oncoming vehicle of the host vehicle is traveling around a curve, even if it is mistakenly determined to be a stopped vehicle and the preceding or oncoming vehicle is determined to be in a state of deviating from the left boundary line or the right boundary line in the image captured by the host vehicle's front camera, by determining whether the captured image includes any of the edges of the four corners of the preceding or oncoming vehicle traveling around the curve that are not shown in the captured image, it is possible to avoid mistakenly recognizing the preceding or oncoming vehicle traveling around the curve as a vehicle subject to steering control and to properly recognize target vehicles that are to be avoided through steering control from the captured image. Effect of the Invention

[0011] According to each aspect of the present invention, a target vehicle that is to be avoided by steering control can be appropriately recognized from a captured image. [Brief description of the drawings]

[0012] [Figure 1] 1 is a block diagram showing a steering control device according to a first embodiment; [Diagram 2] 1A is a diagram showing a first example of a case where a vehicle is recognized as a target vehicle, and FIG. 1B is a diagram showing a first example of a case where a vehicle is not recognized as a target vehicle. [Diagram 3] FIG. 3 is a diagram showing an example of a captured image corresponding to FIG. 2(a). [Figure 4] 13A is a diagram showing a second example of a case where a vehicle is recognized as a target vehicle, and FIG. 13B is a diagram showing a second example of a case where a vehicle is not recognized as a target vehicle. [Diagram 5] 13A is a diagram showing a third example of a case where a vehicle is recognized as a target vehicle, and FIG. 13B is a diagram showing a third example of a case where a vehicle is not recognized as a target vehicle. [Figure 6] 13A is a diagram showing a fourth example of a case where a vehicle is recognized as a target vehicle, and FIG. 13B is a diagram showing a fourth example of a case where a vehicle is not recognized as a target vehicle. [Figure 7] 10 is a flowchart illustrating an example of a target vehicle recognition process. [Figure 8] 13 is a flowchart showing a continuation of the target vehicle recognition process. [Figure 9] FIG. 11 is a block diagram showing a steering control device according to a second embodiment. [Figure 10] 13A is a diagram showing a first example of a case where a vehicle is recognized as a target vehicle in the second embodiment, and FIG. 13B is a diagram showing a first example of a case where a vehicle is not recognized as a target vehicle in the second embodiment. [Figure 11] 13A is a diagram showing a second example of a case where a vehicle is recognized as a target vehicle in the second embodiment, and FIG. 13B is a diagram showing a second example of a case where a vehicle is not recognized as a target vehicle in the second embodiment. [Figure 12] 13A is a diagram showing a third example of a case where a vehicle is recognized as a target vehicle in the second embodiment, and FIG. 13B is a diagram showing a third example of a case where a vehicle is not recognized as a target vehicle in the second embodiment. [Figure 13] 13A is a diagram showing a fourth example of a case where a vehicle is recognized as a target vehicle in the second embodiment, and FIG. 13B is a diagram showing a fourth example of a case where a vehicle is not recognized as a target vehicle in the second embodiment. [Figure 14] 13 is a flowchart illustrating an example of a target vehicle recognition process in a second embodiment. [Figure 15] 10 is a flowchart showing a continuation of the target vehicle recognition process in the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] [First embodiment] As shown in Fig. 1, a steering control device (target vehicle recognition device) 100 according to the first embodiment is a device that is mounted on a vehicle such as a passenger car and executes steering control of the vehicle. Hereinafter, the vehicle on which the steering control device 100 is mounted is referred to as the host vehicle.

[0015] The steering control means, as an example, the control of steering in a PDA [Proactive Driving Assist]. The steering control includes the control of avoiding an avoidance target by steering the vehicle. The steering control controls the steering of the vehicle so that the vehicle passes by the side of a vehicle that is stopped or parked outside the vehicle's driving lane while keeping a distance from the vehicle.

[0016] The steering control device 100 recognizes a target vehicle that is an object to be avoided by steering control. Here, Fig. 2(a) is a diagram showing a first example of a case where a target vehicle is recognized. Fig. 2 shows the host vehicle M, a driving lane R1 in which the host vehicle M is traveling, two boundary lines BL, BR that form the driving lane R1, and another vehicle N1. Also shown is the front left end Nfl1, which is the end on the left side in front of the other vehicle N1, among the four corners (four corner portions in a plan view) of the other vehicle N1. The driving lane R1 is a straight road.

[0017] The other vehicle N1 in FIG. 2(a) is a vehicle stopped (or parked) protruding into the driving lane R1. For example, the other vehicle N1 straddles the boundary line BL of the driving lane R1 and is stopped on the sidewalk beside the lane. In the case of such a vehicle N1, there is a risk that a pedestrian P may jump out in front of the stopped vehicle N1, which is in the blind spot of the host vehicle M. The steering control device 100 uses an image captured by the front camera of the host vehicle M (see FIG. 3) to recognize the other vehicle N1 straddling the boundary line BL as a target vehicle to be avoided by steering control. FIG. 3 and the recognition of the target vehicle will be described in detail later.

[0018] Fig. 2(b) is a diagram showing a first example of a case where the vehicle is not recognized as a target vehicle. Fig. 2(b) shows the driving lane R2 in which the vehicle M is traveling, the other vehicle N2, a virtual projection line BLg of the boundary line BL, and a virtual projection line BRg of the boundary line BR. Also shown is the front left end Nfl2, which is the end on the left side in front of the other vehicle N2, among the four corners (four corners in a plan view) of the other vehicle N2.

[0019] The driving lane R2 is a curve that turns left in the traveling direction of the host vehicle M. The other vehicle N2 is a vehicle traveling on the curve. Although the other vehicle N2 is a traveling vehicle, the accuracy of speed detection decreases as the vehicle moves away from the host vehicle M, so the other vehicle N2 may be erroneously determined to be a stopped vehicle.

[0020] The virtual projection line BLg is a line obtained by projecting the boundary line BL on the captured image captured by the front camera of the vehicle M onto the actual road surface. Similarly, the virtual projection line BRg is a line obtained by projecting the boundary line BR on the captured image captured by the front camera of the vehicle M onto the actual road surface. That is, in the situation shown in FIG. 2(b), the other vehicle N2 turning the curve of the driving lane R2 appears to be straddling the boundary line BL on the captured image captured by the front camera of the vehicle M. In this case, the steering control device 100 uses the fact that the front left end Nfl2 of the other vehicle N2 traveling around the curve is not included in the captured image to prevent the other vehicle N2 from being mistakenly recognized as a target vehicle to be avoided by steering control.

[0021] <Configuration of the steering control device according to the first embodiment> The steering control device 100 includes an ECU [Electronic Control Unit] 10 that performs overall management of the device. The ECU 10 is an electronic control unit having a CPU [Central Processing Unit] and a storage unit such as a ROM [Read Only Memory] or a RAM [Random Access Memory]. In the ECU 10, for example, various functions are realized by the CPU executing a program stored in the storage unit. The ECU 10 may be composed of a plurality of electronic units. In addition, some of the functions of the ECU 10 described below may be executed in a server that can communicate with the vehicle.

[0022] The ECU 10 is connected to a front camera 1, a front radar sensor 2, a vehicle speed sensor 3, an acceleration sensor 4, a yaw rate sensor 5, and an actuator 6. The front camera 1 is a camera for capturing an image of the area in front of the vehicle. The front camera 1 is provided, for example, on the back side of the windshield of the vehicle, and transmits captured images of the area in front of the vehicle to the ECU 10. The front camera 1 may be a monocular camera or a stereo camera.

[0023] The front radar sensor 2 is a detection device that detects an object in front of the vehicle using radio waves (e.g., millimeter waves) or light. The front radar sensor 2 includes, for example, a millimeter wave radar or a LIDAR (Light Detection and Ranging). The front radar sensor 2 detects an object by transmitting radio waves or light to the surroundings of the vehicle and receiving the radio waves or light reflected by the object. The front radar sensor 2 transmits information on the detected object, such as a preceding vehicle, to the ECU 10. The detection result of the front radar sensor 2 also includes speed information of the object (e.g., the relative speed with respect to the vehicle).

[0024] The vehicle speed sensor 3 is a detector that detects the speed of the vehicle. For example, a wheel speed sensor that is provided on the wheels of the vehicle or a drive shaft that rotates together with the wheels and detects the rotation speed of the wheels is used as the vehicle speed sensor 3. The vehicle speed sensor 3 transmits the detected vehicle speed information (wheel speed information) to the ECU 10.

[0025] The acceleration sensor 4 is a detector that detects the acceleration of the host vehicle. The acceleration sensor 4 includes, for example, a longitudinal acceleration sensor that detects the acceleration in the longitudinal direction of the host vehicle, and a lateral acceleration sensor that detects the lateral acceleration of the host vehicle. The acceleration sensor 4 transmits, for example, acceleration information of the host vehicle to the ECU 10.

[0026] The yaw rate sensor 5 is a detector that detects a yaw rate (rotational angular velocity) around a vertical axis of the center of gravity of the vehicle. For example, a gyro sensor can be used as the yaw rate sensor 5. The yaw rate sensor 5 transmits the detected yaw rate information of the vehicle to the ECU 10.

[0027] The actuator 6 is a device used to control the host vehicle. The actuator 6 includes at least a drive actuator, a brake actuator, and a steering actuator. The drive actuator controls the amount of air supplied to the engine (throttle opening) in response to a control signal from the ECU 10, thereby controlling the driving force of the host vehicle. If the host vehicle is a hybrid vehicle (HEV: Hybrid Electric Vehicle), in addition to the amount of air supplied to the engine, a control signal from the ECU 10 is input to a motor serving as a power source to control the driving force. If the host vehicle is an electric vehicle (BEV: Battery Electric Vehicle), a control signal from the ECU 10 is input to a motor serving as a power source to control the driving force. In these cases, the motor serving as a power source constitutes the actuator 6.

[0028] The brake actuator controls a brake system in response to a control signal from the ECU 10, and controls the braking force applied to the wheels of the vehicle. For example, a hydraulic brake system can be used as the brake system. The steering actuator controls the drive of an assist motor that controls the steering torque of the electric power steering system in response to a control signal from the ECU 10. In this way, the steering actuator controls the steering torque of the vehicle.

[0029] Next, a description will be given of the functional configuration of the ECU 10. As shown in Fig. 1, the ECU 10 has a stopped vehicle detection unit 11, a direction determination unit 12, a boundary line departure state determination unit 13, a target vehicle recognition unit 14, and a steering control unit 15. Note that some of the functions of the ECU 10 described below may be executed by a server capable of communicating with the vehicle itself.

[0030] The stopped vehicle detection unit 11 detects a stopped vehicle located in front of the vehicle based on at least one of the captured image of the front camera 1 and the detection result of the front radar sensor 2. A stopped vehicle is a vehicle whose speed is zero. The vehicle is not limited to a four-wheeled vehicle, but may be a two-wheeled vehicle. The method of detecting a stopped vehicle is not particularly limited, and a well-known method can be adopted. The stopped vehicle detection unit 11 may detect a stopped vehicle from, for example, the speed difference between another vehicle in front of the vehicle and a guardrail or other structure in front of the vehicle. The stopped vehicle detection unit 11 may detect a stopped vehicle based on the vehicle speed detected by the vehicle speed sensor 3 of the vehicle. Note that the stopped vehicle detection unit 11 may erroneously detect a traveling vehicle as a stopped vehicle due to an error in speed information caused by sensor accuracy for other distant vehicles.

[0031] When a stopped vehicle is detected by the stopped vehicle detection unit 11, the orientation determination unit 12 determines the orientation of the stopped vehicle based on the image captured by the front camera 1 of the host vehicle. The orientation determination unit 12 determines whether the stopped vehicle is in a forward-facing state with its front facing the host vehicle or in a backward-facing state with its back facing the host vehicle.

[0032] The orientation determination unit 12 determines the orientation of the stopped vehicle from the captured image by image recognition such as pattern matching, deep learning, etc. The orientation determination unit 12 may determine the orientation of the stopped vehicle from the captured image by a known image processing method.

[0033] When a stopped vehicle is detected by the stopped vehicle detection unit 11, the boundary line departure state determination unit 13 determines whether the stopped vehicle is in a right boundary line departure state, a left boundary line departure state, or neither a right boundary line departure state nor a left boundary line departure state, based on an image captured by the front camera of the vehicle. The right boundary line departure state is a state in which the stopped vehicle has straddled the boundary line on the right side of the vehicle's driving lane. The left boundary line departure state is a state in which the stopped vehicle has straddled the boundary line on the left side of the driving lane.

[0034] As an example, the boundary line departure state determination unit 13 performs the above determination by detecting various positions related to the stopped vehicle and the boundary line on the captured image. The boundary line departure state determination unit 13 detects a reference axis that is an axis extending in the horizontal direction of the image from the bottom end of the stopped vehicle on the captured image, a vehicle left end position obtained by projecting the left end of the stopped vehicle on the captured image onto the reference axis, a vehicle right end position obtained by projecting the right end of the stopped vehicle on the captured image onto the reference axis, and a reference boundary line position that is a position where at least one of the two boundaries that form the vehicle's driving lane R1 on the captured image intersects with the reference axis.

[0035] The boundary line deviation state determination unit 13 detects various positions by recognizing stopped vehicles and lane boundaries in the captured image, for example, by deep learning or pattern matching. The boundary line deviation state determination unit 13 may employ other well-known image processing methods.

[0036] Here, Fig. 3 is a diagram showing an example of a captured image corresponding to Fig. 2(a). Fig. 3 shows a captured image G ahead of the vehicle M, a driving lane R1 in the captured image G, two boundary lines BL and BR that form the driving lane R1, a stopped vehicle N1 in the captured image G, a reference axis CL, a vehicle left edge position NL, a vehicle right edge position NR, a first reference boundary line position C1, a second reference boundary line position C2, a first distance DA, and a second distance DB.

[0037] 3, the reference axis CL is an axis extending in the image horizontal direction from the lower end of the stopped vehicle N1 on the captured image G. The image horizontal direction is the horizontal direction in the captured image G. The lower end of the stopped vehicle N1 is, for example, the lower end of the tires of the stopped vehicle N1 on the captured image G.

[0038] The vehicle left end position NL is a position where the left end of the stopped vehicle N1 is projected onto the reference axis line CL on the captured image G. The vehicle right end position NR is a position where the right end of the stopped vehicle N1 is projected onto the reference axis line CL on the captured image G. The left end of the stopped vehicle N1 is the part located at the leftmost end of the stopped vehicle N1. It may be the part of the body of the stopped vehicle N1 that protrudes to the leftmost, or if the stopped vehicle N1 is a cargo vehicle, it may be the part of the loading platform that protrudes to the leftmost. If the left mirror of the stopped vehicle N1 protrudes to the leftmost, the tip of the left mirror may be the left end. The position where the left end of the stopped vehicle N1 is projected perpendicularly to the reference axis line CL is the vehicle left end position NL. The vehicle right end position NR is the same as the vehicle left end position NL, so a description will be omitted.

[0039] The first reference boundary position C1 is the position where the boundary line BL intersects with the reference axis line CL, of the two boundary lines BL, BR that form the vehicle's driving lane R1 on the captured image G. The second reference boundary position C2 is the position where the boundary line BR intersects with the reference axis line CL. When the boundary line BL is composed of multiple pixels in the horizontal direction of the image (when the boundary line BL has a certain width on the captured image G), the first reference boundary position C1 may be the center position of the boundary line BL in the horizontal direction of the image, or may be the end position of the boundary line BL on the vehicle side or the end position opposite the vehicle in the horizontal direction of the image.

[0040] The first distance DA is the distance from the vehicle's own vehicle side, either the vehicle left end position NL or the vehicle right end position NR, on the reference axis line CL to the first reference boundary line position C1. The first distance DA in FIG. 3 is the distance between the first reference boundary line position C1 and the vehicle right end position NR. The second distance DB is the distance from the vehicle's own vehicle side, either the vehicle left end position NL or the vehicle right end position NR, on the reference axis line CL to the first reference boundary line position C1. The second distance DB in FIG. 3 is the distance between the vehicle left end position NL and the first reference boundary line position C1. The boundary line departure state determination unit 13 may calculate the first distance and the second distance corresponding to the second reference boundary line position C2.

[0041] The boundary line departure state determination unit 13 does not necessarily need to detect the first reference boundary line position C1 of the boundary line BL and the first reference boundary line position C2 of the boundary line BR. The boundary line departure state determination unit 13 may be configured to detect only the reference boundary line position C1 corresponding to the boundary line BL that is closer to the stopped vehicle N1, out of the boundary lines BL and BR.

[0042] The boundary line departure state determination unit 13 uses the various detected positions to determine whether or not there is a reference boundary line position (first reference boundary line position C1 or second reference boundary line position C2) sandwiched between the vehicle left end position NL and the vehicle right end position NR of the stopped vehicle N1.

[0043] The boundary line departure state determination unit 13 may determine the horizontal coordinate (coordinate in the horizontal direction of the image) of each position at the pixel level by semantic segmentation in the captured image G, and may make a determination using the difference. For example, if the pixel coordinate of the vehicle left edge position NL is subtracted from the pixel coordinate of the first reference boundary line position C1, the difference will be a negative value when the vehicle left edge position NL is located to the left of the first reference boundary line position C1, and the difference will be a positive value when the vehicle left edge position NL is located to the right of the first reference boundary line position C1.

[0044] Similarly, when the pixel coordinates of the vehicle right edge position NR are subtracted from the pixel coordinates of the first reference boundary position C1, the difference is a negative value when the vehicle right edge position NR is located to the left of the first reference boundary position C1, and the difference is a positive value when the vehicle right edge position NR is located to the right of the first reference boundary position C1. When the positive and negative signs of the difference between the vehicle left edge position NL and the difference between the vehicle right edge position NR are different, the target vehicle recognition unit 14 determines that a first reference boundary position C1 exists that is sandwiched between the vehicle left edge position NL and the vehicle right edge position NR of the stopped vehicle N1.

[0045] The second reference boundary position C2 can be determined in a similar manner. In the situation shown in Fig. 3, the target vehicle recognition unit 14 determines that the second reference boundary position C2 is not sandwiched between the vehicle left end position NL and the vehicle right end position NR because the positive and negative signs of the difference between the second reference boundary position C2 and the vehicle left end position NL are the same as the positive and negative signs of the difference between the second reference boundary position C2 and the vehicle right end position NR. Note that the method of obtaining the image coordinates of each position is not limited to semantic segmentation, and any well-known method can be adopted.

[0046] When the boundary line departure state determination unit 13 determines that a first reference boundary line position C1 exists between the vehicle left end position NL and the vehicle right end position NR of the stopped vehicle N1, the boundary line departure state determination unit 13 determines that the stopped vehicle N1 is in a left boundary line departure state. Similarly, when the boundary line departure state determination unit 13 determines that a second reference boundary line position C2 exists between the vehicle left end position NL and the vehicle right end position NR of the stopped vehicle N1, the boundary line departure state determination unit 13 determines that the stopped vehicle N1 is in neither a left boundary line departure state nor a left boundary line departure state when there is no reference boundary line position between the vehicle left end position NL and the vehicle right end position NR of the stopped vehicle N1.

[0047] In addition, the boundary line departure state determination unit 13 may determine that the other vehicle N2 is in a left boundary line departure state when the other vehicle N2 in a curve is located on a virtual projection line BLg of the boundary line BL (a line projected onto the actual road surface of the boundary line BL in the captured image), as shown in Figure 2 (b).

[0048] When a stopped vehicle is detected by the stopped vehicle detection unit 11, the target vehicle recognition unit 14 recognizes the target vehicle for steering control based on the image captured by the front camera 1 of the vehicle, the judgment result of the orientation judgment unit 12, and the judgment result of the boundary line departure state judgment unit 13.

[0049] In addition, when the boundary line departure state determination unit 13 determines that the stopped vehicle is not in a right boundary line departure state or a left boundary line departure state, the target vehicle recognition unit 14 does not recognize the stopped vehicle as a target vehicle for steering control. In other words, the target vehicle recognition unit 14 does not recognize a vehicle that is not stopped straddling the boundary line BL or boundary line BR as a target vehicle.

[0050] When the orientation determination unit 12 determines that the stopped vehicle is facing backward and the boundary line departure state determination unit 13 determines that the stopped vehicle is in a left boundary line departure state, and the front left end of the stopped vehicle is not included in the captured image, the target vehicle recognition unit 14 recognizes the stopped vehicle as a target vehicle for steering control.

[0051] The front left edge is the front left edge of the four corners (four corners in a plan view) of the stopped vehicle. The remaining are the front right edge, rear left edge, and rear right edge. Based on the orientation of the stopped vehicle determined by the orientation determination unit 12, the target vehicle recognition unit 14 determines whether or not the front left edge of the stopped vehicle is included in the captured image by image recognition of the captured image using pattern matching, deep learning, or the like. The same applies to the determination of the front right edge, rear left edge, and rear right edge.

[0052] Specifically, in the situation shown in Fig. 2(a), it is determined that the stopped vehicle N1 is facing backward and deviating from the left boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the front left edge Nfl1 of the stopped vehicle N1 is included in the captured image G. In the situation shown in Fig. 2(a), since the front left edge Nfl1 of the stopped vehicle N1 is not included in the captured image G as seen from the host vehicle M, the target vehicle recognition unit 14 recognizes the stopped vehicle N1 as a target vehicle.

[0053] On the other hand, when the front left edge of the stopped vehicle is included in the captured image, the target vehicle recognition unit 14 does not recognize the stopped vehicle as a target vehicle for steering control. Specifically, in the situation shown in Fig. 2(b), the other vehicle N2 is determined to be a stopped vehicle by the stopped vehicle detection unit 11 due to a speed information error. In the following, the other vehicle N2 traveling around a curve will be described as the stopped vehicle N2.

[0054] In the situation shown in FIG. 2(b), the stopped vehicle N2 in the curve is located on a virtual projection line BLg of the boundary line BL (a line obtained by projecting the boundary line BL on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing backward and deviating from the left boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the front left edge Nfl2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 2(b), the front left edge Nfl2 of the stopped vehicle N2 is not included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 14 does not recognize the stopped vehicle N2 as a target vehicle.

[0055] Fig. 4(a) is a diagram showing a second example of a case where a stopped vehicle N1 is recognized as a target vehicle. In Fig. 4(a), the stopped vehicle N1 is different from Fig. 2(a) in that the stopped vehicle N1 faces forward with its front facing the host vehicle M. Fig. 4(a) shows the rear right end Nrr1 of the stopped vehicle N1.

[0056] In the situation shown in Fig. 4(a), the stopped vehicle N1 is determined to be facing forward and deviating from the left boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the rear right end Nrr1 of the stopped vehicle N1 is included in the captured image G. Since the rear right end Nrr1 of the stopped vehicle N1 is included in the captured image G as seen from the host vehicle M in the situation shown in Fig. 4(a), the target vehicle recognition unit 14 recognizes the stopped vehicle N1 as a target vehicle.

[0057] Fig. 4(b) is a diagram showing a second example of a case where the stopped vehicle N2 is not recognized as a target vehicle. In Fig. 4(b), the stopped vehicle N2 is traveling on the outside of the driving lane R2 (for example, the oncoming lane) and is facing forward with its front facing the host vehicle M, which is different from Fig. 2(b). Fig. 4(b) shows the rear right edge Nrr2 of the stopped vehicle N2.

[0058] In the situation shown in FIG. 4(b), the stopped vehicle N2 is located on a virtual projection line BLg of the boundary line BL (a line obtained by projecting the boundary line BL on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing forward and deviating from the left boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the rear right end Nrr2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 4(b), the target vehicle recognition unit 14 does not recognize the stopped vehicle N2 as a target vehicle because the rear right end Nrr2 of the stopped vehicle N2 is included in the captured image G as seen from the host vehicle M.

[0059] Fig. 5(a) is a diagram showing a third example of a case where a vehicle is recognized as a target vehicle. Fig. 5(a) is different from Fig. 2(a) in that the stopped vehicle N1 is located to the right of the host vehicle M and deviates from the right boundary line. Fig. 5(a) shows the front right edge Nfr1 of the stopped vehicle N1.

[0060] In the situation shown in Fig. 5(a), the stopped vehicle N1 is determined to be facing backward and deviating from the right boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the front right edge Nfr1 of the stopped vehicle N1 is included in the captured image G. In the situation shown in Fig. 5(a), the front right edge Nfr1 of the stopped vehicle N1 is not included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 14 recognizes the stopped vehicle N1 as a target vehicle.

[0061] Fig. 5(b) is a diagram showing a third example of a case where the vehicle is not recognized as a target vehicle. In Fig. 5(b), the host vehicle M and the stopped vehicle N2 are traveling in a driving lane R3 that turns to the right, which is different from Fig. 2(b). Fig. 5(b) shows the front right edge Nfr2 of the stopped vehicle N2.

[0062] In the situation shown in FIG. 5(b), the stopped vehicle N2 in the curve is located on a virtual projection line BRg of the boundary line BR (a line obtained by projecting the boundary line BR on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing backward and deviating from the right boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the front right end Nfr2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 5(b), the front right end Nfr2 of the stopped vehicle N2 is included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 14 does not recognize the stopped vehicle N2 as a target vehicle.

[0063] Fig. 6(a) is a diagram showing a fourth example of a case where a stopped vehicle N1 is recognized as a target vehicle. In Fig. 6(a), the stopped vehicle N1 faces forward toward the host vehicle M and is in a state of deviating from the right boundary line located to the right of the host vehicle M, which is different from Fig. 2(a). Fig. 6(a) shows the rear left edge Nfl1 of the stopped vehicle N1.

[0064] In the situation shown in Fig. 6(a), the stopped vehicle N1 is determined to be facing forward and deviating from the right boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the rear left edge Nfl1 of the stopped vehicle N1 is included in the captured image G. Since the rear left edge Nfl1 of the stopped vehicle N1 is not included in the captured image G as seen from the host vehicle M in the situation shown in Fig. 6(a), the target vehicle recognition unit 14 recognizes the stopped vehicle N1 as a target vehicle.

[0065] Fig. 6(b) is a diagram showing a fourth example of a case where the vehicle is not recognized as a target vehicle. In Fig. 6(b), the host vehicle M is traveling in a driving lane R3 that turns to the right, and the stopped vehicle N2 is traveling on the outside of the driving lane R2 (such as another road) and facing forward with its front facing the host vehicle M, which is different from Fig. 2(b). Fig. 6(b) shows the rear left edge Nrl2 of the stopped vehicle N2.

[0066] In the situation shown in FIG. 6(b), the stopped vehicle N2 is located on a virtual projection line BRg of the boundary line BR (a line obtained by projecting the boundary line BR on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing forward and deviating from the right boundary line. In this case, the target vehicle recognition unit 14 determines whether or not the rear left end Nrl2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 6(b), the target vehicle recognition unit 14 does not recognize the stopped vehicle N2 as a target vehicle because the rear left end Nrl2 of the stopped vehicle N2 is included in the captured image G as seen from the host vehicle M.

[0067] 3 is greater than the second distance DB, the target vehicle recognition unit 14 may not recognize the stopped vehicle N1 as a target vehicle even if the end of the stopped vehicle N1 used for the above determination is not included in the captured image G. That is, since it is highly likely that a vehicle preceding the host vehicle is temporarily stopped when most of the stopped vehicle N1 is located on the driving lane R1 of the host vehicle, the target vehicle recognition unit 14 does not recognize the stopped vehicle N1 as a target vehicle for steering control when the first distance DA on the host vehicle side on the reference axis line CL is greater than the second distance DB, thereby making it possible to avoid erroneously setting the temporarily stopped vehicle preceding the host vehicle as an avoidance target for steering control.

[0068] When a target vehicle that is an avoidance target of steering control is recognized by the target vehicle recognition unit 14, the steering control unit 15 executes steering control to avoid the target vehicle and pass the side of the target vehicle. The steering control unit 15 executes steering control based on the image captured by the front camera 1 and / or the detection result of the front radar sensor 2, taking into account the vehicle speed, acceleration, and yaw rate of the host vehicle. Note that the steering control unit 15 does not necessarily need to execute steering control even when a target vehicle is recognized.

[0069] <Processing method of steering control device according to the first embodiment> Next, a processing method of the steering control device (target vehicle recognition device) 100 according to this embodiment will be described. Fig. 7 is a flowchart showing an example of the target vehicle recognition process. Fig. 8 is a flowchart showing a continuation of the target vehicle recognition process. The target vehicle recognition process is performed, for example, when the driver's setting of whether or not to execute steering control of the vehicle is set to "necessary."

[0070] 7, the ECU 10 of the steering control device 100 detects a stopped vehicle N1 located in front of the host vehicle by the stopped vehicle detection unit 11 as S10 (stopped vehicle detection step). The stopped vehicle detection unit 11 detects the stopped vehicle N1 based on at least one of an image captured by the front camera 1 and a detection result of the front radar sensor 2.

[0071] In S11, the ECU 10 performs image recognition of the captured image using the orientation determination unit 12 and the boundary line departure state determination unit 13 (image recognition step). The orientation determination unit 12 performs image recognition of the captured image using, for example, deep learning or pattern matching. The boundary line departure state determination unit 13 similarly detects various positions of the stopped vehicle and boundary line in the captured image using image recognition.

[0072] In S12, the ECU 10 determines whether the stopped vehicle is facing forward or backward by the direction determination unit 12 (direction determination step). If the ECU 10 determines that the stopped vehicle is facing backward (S12: YES), the ECU 10 proceeds to S13. If the ECU 10 determines that the stopped vehicle is not facing backward (facing forward) (S12: NO), the ECU 10 proceeds to S20 in the flowchart of FIG. 8.

[0073] In S13, the ECU 10 determines whether the stopped vehicle is in a left boundary line departure state by the boundary line departure state determination unit 13 (first boundary line departure state determination step). If the ECU 10 determines that the stopped vehicle is in a left boundary line departure state (S13: YES), the ECU 10 proceeds to S14. If the ECU 10 does not determine that the stopped vehicle is in a left boundary line departure state (S13: NO), the ECU 10 proceeds to S15.

[0074] In S14, the ECU 10 determines whether or not the front left edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 14 (front left edge determination step). If the ECU 10 determines that the front left edge of the stopped vehicle is included in the captured image (S14: YES), the ECU 10 proceeds to S17. If the ECU 10 does not determine that the front left edge of the stopped vehicle is included in the captured image (S14: NO), the ECU 10 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0075] In S15, the ECU 10 determines whether the stopped vehicle is in a right boundary line departure state by the boundary line departure state determination unit 13 (second boundary line departure state determination step). If the ECU 10 determines that the stopped vehicle is in a right boundary line departure state (S15: YES), the ECU 10 proceeds to S16. If the ECU 10 does not determine that the stopped vehicle is in a right boundary line departure state (S15: NO), the ECU 10 determines that the stopped vehicle is neither in a left boundary line departure state nor in a right boundary line departure state, and ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0076] In S16, the ECU 10 determines whether or not the front right edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 14 (front right edge determination step). If the ECU 10 determines that the front right edge of the stopped vehicle is included in the captured image (S16: YES), the ECU 10 proceeds to S17. If the ECU 10 does not determine that the front right edge of the stopped vehicle is included in the captured image (S16: NO), the ECU 10 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0077] In S17, the ECU 10 recognizes the stopped vehicle as a target vehicle to be avoided by steering control using the target vehicle recognition unit 14 (target vehicle recognition step). After that, the ECU 10 ends the target vehicle recognition process.

[0078] 8, in S20, the ECU 10 determines whether the stopped vehicle is in a left boundary line departure state by the boundary line departure state determination unit 13 (third boundary line departure state determination step). If the ECU 10 determines that the stopped vehicle is in a left boundary line departure state (S20: YES), the ECU 10 proceeds to S21. If the ECU 10 does not determine that the stopped vehicle is in a left boundary line departure state (S20: NO), the ECU 10 proceeds to S22.

[0079] In S21, the ECU 10 determines whether or not the rear right edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 14 (rear right edge determination step). If the ECU 10 determines that the rear right edge of the stopped vehicle is included in the captured image (S21: YES), the ECU 10 proceeds to S24. If the ECU 10 does not determine that the rear right edge of the stopped vehicle is included in the captured image (S21: NO), the ECU 10 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0080] In S22, the ECU 10 determines whether the stopped vehicle is in a right boundary line departure state by the boundary line departure state determination unit 13 (fourth boundary line departure state determination step). If the ECU 10 determines that the stopped vehicle is in a right boundary line departure state (S22: YES), the ECU 10 proceeds to S23. If the ECU 10 does not determine that the stopped vehicle is in a right boundary line departure state (S22: NO), the ECU 10 determines that the stopped vehicle is neither in a left boundary line departure state nor in a right boundary line departure state, and ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0081] In S23, the ECU 10 determines whether or not the rear left edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 14 (rear left edge determination step). If the ECU 10 determines that the rear left edge of the stopped vehicle is included in the captured image (S23: YES), the ECU 10 proceeds to S24. If the ECU 10 does not determine that the rear left edge of the stopped vehicle is included in the captured image (S23: NO), the ECU 10 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0082] In S24, the ECU 10 recognizes the stopped vehicle as a target vehicle to be avoided by steering control using the target vehicle recognition unit 14. After that, the ECU 10 ends the target vehicle recognition process.

[0083] According to the steering control device 100 of the first embodiment described above, when a vehicle preceding the vehicle or an oncoming vehicle is traveling around a curve, even if the vehicle is mistakenly determined to be a stopped vehicle and the preceding vehicle or the oncoming vehicle is determined to be in a state of deviating from the left boundary line or the right boundary line in the image captured by the front camera of the vehicle, by determining whether the captured image includes any of the edges of the four corners of the preceding vehicle or the oncoming vehicle traveling around the curve that are not shown in the captured image, it is possible to avoid mistakenly recognizing the preceding vehicle or the oncoming vehicle traveling around the curve as a vehicle subject to steering control and to properly recognize target vehicles that are to be avoided by steering control from the captured image.

[0084] [Second embodiment] Next, a steering control device (target vehicle recognition device) according to a second embodiment will be described with reference to the drawings. Fig. 9 is a block diagram showing the steering control device according to the second embodiment.

[0085] <Configuration of the steering control device according to the second embodiment> A steering control device 200 according to the second embodiment shown in Fig. 9 differs from the first embodiment only in the function of the target vehicle recognition unit 21 in the ECU 20. The target vehicle recognition unit 21 according to the second embodiment recognizes the target vehicle by focusing on an end portion that is included in the captured image if the vehicle is the target, rather than an end portion that is not included in the captured image if the vehicle is the target, among the four corners of the stopped vehicle.

[0086] In other words, when the orientation determination unit 12 determines that the stopped vehicle is facing backward and the boundary line departure state determination unit 13 determines that the stopped vehicle is in a left boundary line departure state, the target vehicle recognition unit 21 recognizes the stopped vehicle as a target vehicle for steering control when the front right end of the stopped vehicle is included in the captured image.

[0087] Fig. 10(a) is a diagram showing a first example of a case where a vehicle is recognized as a target vehicle in the second embodiment. In the situation shown in Fig. 10(a), the stopped vehicle N1 is determined to be facing backward and deviating from the left boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the front right end Nfr1 of the stopped vehicle N1 is included in the captured image G. In the situation shown in Fig. 10(a), the front right end Nfr1 of the stopped vehicle N1 is included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 21 recognizes the stopped vehicle N1 as a target vehicle.

[0088] On the other hand, when the front right edge of the stopped vehicle is not included in the captured image, the target vehicle recognition unit 21 does not recognize the stopped vehicle as a target vehicle for steering control. Here, Fig. 10(b) is a diagram showing a first example of a case where the vehicle is not recognized as a target vehicle in the second embodiment. In the situation shown in Fig. 10(b), it is assumed that the other vehicle N2 is determined to be a stopped vehicle by the stopped vehicle detection unit 11 due to a speed information error.

[0089] In the situation shown in FIG. 10(b), the stopped vehicle N2 in the curve is located on a virtual projection line BLg of the boundary line BL (a line obtained by projecting the boundary line BL on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing backward and deviating from the left boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the front right end Nfr2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 10(b), the front right end Nfr2 of the stopped vehicle N2 is not included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 21 does not recognize the stopped vehicle N2 as a target vehicle.

[0090] Fig. 11(a) is a diagram showing a second example of a case where a stopped vehicle N1 is recognized as a target vehicle in the second embodiment. In Fig. 11(a), the stopped vehicle N1 is facing forward with its front facing the host vehicle M, which is different from Fig. 10(a).

[0091] In the situation shown in Fig. 11(a), the stopped vehicle N1 is determined to be facing forward and deviating from the left boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the rear left end Nrl1 of the stopped vehicle N1 is included in the captured image G. Since the rear left end Nrl1 of the stopped vehicle N1 is included in the captured image G as seen from the host vehicle M in the situation shown in Fig. 11(a), the target vehicle recognition unit 21 recognizes the stopped vehicle N1 as a target vehicle.

[0092] Fig. 11(b) is a diagram showing a second example of a case where a stopped vehicle N2 is not recognized as a target vehicle in the second embodiment. In Fig. 11(b), the stopped vehicle N2 is traveling on the outside of the traveling lane R2 (for example, the oncoming lane) and is facing forward with its front facing the host vehicle M, which is different from Fig. 10(b).

[0093] In the situation shown in FIG. 11(b), since the stopped vehicle N2 is located on a virtual projection line BLg of the boundary line BL (a line obtained by projecting the boundary line BL on the captured image onto the actual road surface), the stopped vehicle N2 is determined to be facing forward and deviating from the left boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the rear left end Nrl2 of the stopped vehicle N2 is included in the captured image G. Since the rear left end Nrl2 of the stopped vehicle N2 is not included in the captured image G as seen from the host vehicle M in the situation shown in FIG. 11(b), the target vehicle recognition unit 21 does not recognize the stopped vehicle N2 as a target vehicle.

[0094] Fig. 12(a) is a diagram showing a third example of a case where a stopped vehicle N1 is recognized as a target vehicle in the second embodiment. In Fig. 12(a), the stopped vehicle N1 is located to the right of the host vehicle M and is in a state of deviating from the right boundary line, which is different from Fig. 10(a).

[0095] In the situation shown in Fig. 12(a), the stopped vehicle N1 is determined to be facing backward and deviating from the right boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the front left edge Nfl1 of the stopped vehicle N1 is included in the captured image G. Since the front left edge Nfl1 of the stopped vehicle N1 is included in the captured image G as seen from the host vehicle M in the situation shown in Fig. 12(a), the target vehicle recognition unit 21 recognizes the stopped vehicle N1 as a target vehicle.

[0096] Fig. 12(b) is a diagram showing a third example of a case where the vehicle is not recognized as a target vehicle in the second embodiment. In Fig. 12(b), the host vehicle M and the stopped vehicle N2 are traveling in a driving lane R3 that turns to the right, which is different from Fig. 10(b).

[0097] In the situation shown in FIG. 12(b), the stopped vehicle N2 in the curve is located on a virtual projection line BRg of the boundary line BR (a line obtained by projecting the boundary line BR on the captured image onto the actual road surface), so it is determined that the stopped vehicle N2 is facing backward and deviating from the right boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the front left edge Nfl2 of the stopped vehicle N2 is included in the captured image G. In the situation shown in FIG. 12(b), the front left edge Nfl2 of the stopped vehicle N2 is not included in the captured image G as seen from the host vehicle M, so the target vehicle recognition unit 21 does not recognize the stopped vehicle N2 as a target vehicle.

[0098] Fig. 13(a) is a diagram showing a fourth example of a case where a stopped vehicle N1 is recognized as a target vehicle in the second embodiment. In Fig. 13(a), the stopped vehicle N1 faces forward with its front facing the host vehicle M, and is located to the right of the host vehicle M in a state of deviating from the right boundary line, which is different from Fig. 10(a).

[0099] In the situation shown in Fig. 13(a), the stopped vehicle N1 is determined to be facing forward and deviating from the right boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the rear right end Nrr1 of the stopped vehicle N1 is included in the captured image G. Since the rear right end Nrr1 of the stopped vehicle N1 is included in the captured image G as seen from the host vehicle M in the situation shown in Fig. 13(a), the target vehicle recognition unit 21 recognizes the stopped vehicle N1 as a target vehicle.

[0100] Fig. 13(b) is a diagram showing a fourth example of a case where the vehicle is not recognized as a target vehicle in the second embodiment. In Fig. 13(b), the host vehicle M is traveling in a driving lane R3 that turns to the right, and the stopped vehicle N2 is traveling on the outside of the driving lane R2 (such as another road) and facing forward with its front side facing the host vehicle M, which is different from Fig. 10(b).

[0101] In the situation shown in FIG. 13(b), since the stopped vehicle N2 is located on a virtual projection line BRg of the boundary line BR (a line obtained by projecting the boundary line BR on the captured image onto the actual road surface), it is determined that the stopped vehicle N2 is facing forward and deviating from the right boundary line. In this case, the target vehicle recognition unit 21 determines whether or not the rear right end Nrr2 of the stopped vehicle N2 is included in the captured image G. Since the rear right end Nrr2 of the stopped vehicle N2 is not included in the captured image G as seen from the host vehicle M in the situation shown in FIG. 13(b), the target vehicle recognition unit 21 does not recognize the stopped vehicle N2 as a target vehicle.

[0102] <Processing method of steering control device according to second embodiment> Next, a processing method of the steering control device (target vehicle recognition device) 200 according to the second embodiment will be described. Fig. 14 is a flowchart showing an example of the target vehicle recognition process in the second embodiment. Fig. 15 is a flowchart showing a continuation of the target vehicle recognition process in the second embodiment. The target vehicle recognition process is performed, for example, when the driver is set to execute steering control of the host vehicle.

[0103] 14, the ECU 20 of the steering control device 200 detects a stopped vehicle N1 located in front of the host vehicle by the stopped vehicle detection unit 11 as S30 (stopped vehicle detection step). The stopped vehicle detection unit 11 detects the stopped vehicle N1 based on at least one of the captured image of the front camera 1 and the detection result of the front radar sensor 2.

[0104] In S31, the ECU 20 performs image recognition of the captured image by the orientation determination unit 12 and the boundary line departure state determination unit 13 (image recognition step). The orientation determination unit 12 performs image recognition of the captured image by, for example, deep learning or pattern matching. The boundary line departure state determination unit 13 similarly detects various positions of the stopped vehicle and boundary line in the captured image by image recognition.

[0105] In S32, the ECU 20 determines whether the stopped vehicle is facing forward or backward by the direction determination unit 12 (direction determination step). If the ECU 20 determines that the stopped vehicle is facing backward (S32: YES), the process proceeds to S33. If the ECU 20 determines that the stopped vehicle is not facing backward (facing forward) (S32: NO), the process proceeds to S40 in the flowchart of FIG.

[0106] In S33, the ECU 20 determines whether the stopped vehicle is in a left boundary line departure state by the boundary line departure state determination unit 13 (first boundary line departure state determination step). If the ECU 20 determines that the stopped vehicle is in a left boundary line departure state (S33: YES), the ECU 20 proceeds to S34. If the ECU 20 does not determine that the stopped vehicle is in a left boundary line departure state (S33: NO), the ECU 20 proceeds to S35.

[0107] In S34, the ECU 20 determines whether or not the front right edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 21 (front right edge determination step). If the ECU 20 determines that the front right edge of the stopped vehicle is included in the captured image (S34: YES), the ECU 20 proceeds to S37. If the ECU 20 does not determine that the front right edge of the stopped vehicle is included in the captured image (S34: NO), the ECU 20 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0108] In S35, the ECU 20 determines whether the stopped vehicle is in a right boundary line departure state by the boundary line departure state determination unit 13 (second boundary line departure state determination step). If the ECU 20 determines that the stopped vehicle is in a right boundary line departure state (S35: YES), the ECU 20 proceeds to S36. If the ECU 20 does not determine that the stopped vehicle is in a right boundary line departure state (S35: NO), the ECU 20 determines that the stopped vehicle is neither in a left boundary line departure state nor in a right boundary line departure state, and ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0109] In S36, the ECU 20 determines whether or not the front left edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 21 (front left edge determination step). If the ECU 20 determines that the front left edge of the stopped vehicle is included in the captured image (S36: YES), the ECU 20 proceeds to S37. If the ECU 20 does not determine that the front left edge of the stopped vehicle is included in the captured image (S36: NO), the ECU 20 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0110] In S37, the ECU 20 recognizes the stopped vehicle as a target vehicle to be avoided by steering control using the target vehicle recognition unit 21 (target vehicle recognition step). After that, the ECU 20 ends the target vehicle recognition process.

[0111] 8, in S40, the ECU 20 determines whether the stopped vehicle is in a left boundary line departure state by the boundary line departure state determination unit 13 (third boundary line departure state determination step). If the ECU 20 determines that the stopped vehicle is in a left boundary line departure state (S40: YES), the ECU 20 proceeds to S41. If the ECU 20 does not determine that the stopped vehicle is in a left boundary line departure state (S40: NO), the ECU 20 proceeds to S42.

[0112] In S41, the ECU 20 determines whether or not the rear left edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 21 (rear left edge determination step). If the ECU 20 determines that the rear left edge of the stopped vehicle is included in the captured image (S41: YES), the ECU 20 proceeds to S44. If the ECU 20 does not determine that the rear left edge of the stopped vehicle is included in the captured image (S41: NO), the ECU 20 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0113] In S42, the ECU 20 determines whether the stopped vehicle is in a right boundary line departure state by the boundary line departure state determination unit 13 (fourth boundary line departure state determination step). If the ECU 20 determines that the stopped vehicle is in a right boundary line departure state (S42: YES), the ECU 20 proceeds to S43. If the ECU 20 does not determine that the stopped vehicle is in a right boundary line departure state (S42: NO), the ECU 20 determines that the stopped vehicle is neither in a left boundary line departure state nor in a right boundary line departure state, and ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0114] In S43, the ECU 20 determines whether or not the rear right edge of the stopped vehicle is included in the captured image by the target vehicle recognition unit 21 (rear right edge determination step). If the ECU 20 determines that the rear right edge of the stopped vehicle is included in the captured image (S43: YES), the ECU 20 proceeds to S44. If the ECU 20 does not determine that the rear right edge of the stopped vehicle is included in the captured image (S43: NO), the ECU 20 ends the target vehicle recognition process without recognizing the stopped vehicle as a target vehicle.

[0115] In S44, the ECU 20 recognizes the stopped vehicle as a target vehicle to be avoided by steering control using the target vehicle recognition unit 21. After that, the ECU 20 ends the target vehicle recognition process.

[0116] Even in the steering control device 200 of the second embodiment described above, when a vehicle preceding the vehicle or an oncoming vehicle is traveling around a curve, even if it is erroneously determined to be a stopped vehicle and the preceding vehicle or the oncoming vehicle is determined to be in a state of deviating from the left boundary line or the right boundary line in the image captured by the front camera of the vehicle, by determining whether or not the captured image includes any of the edges of the four corners of the preceding vehicle or the oncoming vehicle traveling around the curve that are not shown in the captured image, it is possible to avoid erroneously recognizing the preceding vehicle or the oncoming vehicle traveling around the curve as a vehicle subject to steering control, and to properly recognize target vehicles that are to be avoided by steering control from the captured image.

[0117] Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments. The present disclosure can be implemented in various forms including the above-described embodiments and various modifications and improvements based on the knowledge of those skilled in the art.

[0118] The target vehicle recognition device according to the present disclosure does not necessarily have to be integrated with the steering control device. The target vehicle recognition device may be separate from the steering control device, and some of its functions may be executed by a server capable of communicating with the host vehicle. In this case, the host vehicle does not necessarily have to be a vehicle equipped with all the configurations of the target vehicle recognition device, but means a vehicle that executes steering control.

[0119] The boundary line departure state determination section 13 does not necessarily need to determine the boundary line departure state using each pixel coordinate of the captured image. The boundary line departure state determination section 13 can determine the boundary line departure state using various well-known methods. The boundary line departure state determination section 13 may perform the determination using the detection result of the forward radar sensor 2. In this case, the boundary line departure state determination section 13 may recognize not only other vehicles but also boundary lines (white lines) from the detection result of the forward radar sensor 2.

[0120] The target vehicle recognition unit 21 in the second embodiment may recognize the target vehicle in consideration of the first distance DA and the second distance DB, similar to the target vehicle recognition unit 14 in the first embodiment. [Explanation of symbols]

[0121] 1...forward camera, 2...forward radar sensor, 10, 20...ECU, 11...stopped vehicle detection unit, 12...orientation determination unit, 13...boundary line departure state determination unit, 14, 21...target vehicle recognition unit, 100, 200...steering control device (target vehicle recognition device), BL, BR...boundary line, G...captured image, M...own vehicle, N1, N2...stopped vehicles.

Claims

1. A target vehicle recognition device for recognizing a target vehicle that is an avoidance target of steering control of a host vehicle, a stopped vehicle detection unit that detects a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle; an orientation determination unit that determines whether the stopped vehicle is in a forward facing state with its front facing the host vehicle or in a backward facing state with its back facing the host vehicle based on an image captured by a front camera of the host vehicle; a boundary line departure state determination unit that determines, based on an image captured by a front camera of the host vehicle, whether the stopped vehicle is in a right boundary line departure state crossing a boundary line on the right side of the vehicle's driving lane, whether the stopped vehicle is in a left boundary line departure state crossing a boundary line on the left side of the driving lane, or whether the stopped vehicle is in neither of the right boundary line departure state nor the left boundary line departure state; a target vehicle recognition unit that recognizes the target vehicle of the steering control based on an image captured by a front camera of the host vehicle, a determination result of the orientation determination unit, and a determination result of the boundary line departure state determination unit; Equipped with the target vehicle recognition unit does not recognize the stopped vehicle as the target vehicle of the steering control when the boundary line departure state determination unit determines that the stopped vehicle is not in either the right boundary line departure state or the left boundary line departure state, when the orientation determination unit determines that the stopped vehicle is in the rear-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the left boundary line departure state, if a front left edge of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a front left edge of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, when the orientation determination unit determines that the stopped vehicle is in the rear-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the right boundary line departure state, if a front right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a front right end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, when the orientation determination unit determines that the stopped vehicle is in the forward-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the left boundary line departure state, if a rear right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a rear right end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, A target vehicle recognition device which, when the orientation determination unit determines that the stopped vehicle is in the forward-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the right boundary line departure state, recognizes the stopped vehicle as the target vehicle for the steering control when the rear left end of the stopped vehicle is not included in the captured image, and does not recognize the stopped vehicle as the target vehicle for the steering control when the rear left end of the stopped vehicle is included in the captured image.

2. A target vehicle recognition device for recognizing a target vehicle that is an avoidance target of steering control of a host vehicle, a stopped vehicle detection unit that detects a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle; an orientation determination unit that determines whether the stopped vehicle is in a forward facing state with its front facing the host vehicle or in a backward facing state with its back facing the host vehicle based on an image captured by a front camera of the host vehicle; a boundary line departure state determination unit that determines, based on an image captured by a front camera of the host vehicle, whether the stopped vehicle is in a right boundary line departure state crossing a boundary line on the right side of the vehicle's driving lane, whether the stopped vehicle is in a left boundary line departure state crossing a boundary line on the left side of the driving lane, or whether the stopped vehicle is in neither of the right boundary line departure state nor the left boundary line departure state; a target vehicle recognition unit that recognizes the target vehicle of the steering control based on an image captured by a front camera of the host vehicle, a determination result of the orientation determination unit, and a determination result of the boundary line departure state determination unit; Equipped with the target vehicle recognition unit does not recognize the stopped vehicle as the target vehicle of the steering control when the boundary line departure state determination unit determines that the stopped vehicle is not in either the right boundary line departure state or the left boundary line departure state, when the orientation determination unit determines that the stopped vehicle is in the rear-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the left boundary line departure state, if a front right end of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, and when a front right end of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, when the orientation determination unit determines that the stopped vehicle is in the rear-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the right boundary line departure state, if a front left edge of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, and when a front left edge of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, when the orientation determination unit determines that the stopped vehicle is in the forward-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the left boundary line departure state, if a rear left end of the stopped vehicle is not included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, and when a rear left end of the stopped vehicle is included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, a target vehicle recognition device which, when the orientation determination unit determines that the stopped vehicle is in the forward-facing state and the boundary line departure state determination unit determines that the stopped vehicle is in the right boundary line departure state, does not recognize the stopped vehicle as the target vehicle for the steering control if the rear right end of the stopped vehicle is not included in the captured image, and recognizes the stopped vehicle as the target vehicle for the steering control if the rear right end of the stopped vehicle is included in the captured image.

3. A processing method for a target vehicle recognition device for recognizing a target vehicle that is an avoidance target of steering control of a host vehicle, comprising: a stopped vehicle detection step of detecting a stopped vehicle located in front of the host vehicle based on at least one of an image captured by a front camera of the host vehicle and a detection result of a front radar sensor of the host vehicle; an orientation determination step of determining whether the stopped vehicle is in a forward facing state with its front facing the host vehicle or in a backward facing state with its back facing the host vehicle based on an image captured by a front camera of the host vehicle; a boundary line departure state determination step of determining whether the stopped vehicle is in a right boundary line departure state crossing the boundary line on the right side of the driving lane of the vehicle, whether the stopped vehicle is in a left boundary line departure state crossing the boundary line on the left side of the driving lane, or whether the stopped vehicle is in neither the right boundary line departure state nor the left boundary line departure state based on an image captured by a front camera of the vehicle; a target vehicle recognition step of recognizing a target vehicle of the steering control based on an image captured by a front camera of the host vehicle, a determination result of the direction determination step, and a determination result of the boundary line departure state determination step; Including, In the target vehicle recognition step, when it is determined that the stopped vehicle is not in the right boundary line departure state or the left boundary line departure state in the boundary line departure state determination step, the stopped vehicle is not recognized as the target vehicle of the steering control, When the stopped vehicle is determined to be in the rear-facing state in the direction determination step and when the stopped vehicle is determined to be in the left boundary line departure state in the boundary line departure state determination step, if a front left end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a front left end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, When the stopped vehicle is determined to be in the rear-facing state in the direction determination step and when the stopped vehicle is determined to be in the right boundary line departure state in the boundary line departure state determination step, if a front right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a front right end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, When the stopped vehicle is determined to be in the forward-facing state in the direction determination step and when the stopped vehicle is determined to be in the left boundary line departure state in the boundary line departure state determination step, if a rear right end of the stopped vehicle is not included in the captured image, the stopped vehicle is recognized as a target vehicle for the steering control, and when a rear right end of the stopped vehicle is included in the captured image, the stopped vehicle is not recognized as a target vehicle for the steering control, A processing method for a target vehicle recognition device, comprising: when the stopped vehicle is determined to be in the forward-facing state in the orientation determination step and when the stopped vehicle is determined to be in the right boundary line departure state determination step, if the rear left end of the stopped vehicle is not included in the captured image, recognizing the stopped vehicle as the target vehicle for the steering control; and when the rear left end of the stopped vehicle is included in the captured image, not recognizing the stopped vehicle as the target vehicle for the steering control.

Citation Information

Patent Citations

  • Method of manufacturing ornamental board

    JP1977004525A

  • Vehicle detector, vehicle detection program, and vehicle detection method

    JP2011048420A

  • Safety driving support device and safety driving support method

    JP2016018354A

  • Object recognition method and object recognition system

    JP2021068316A