Driving assistance device, driving assistance method, and program

Abstract

To prevent unnecessary actuation of driving force suppression control.SOLUTION: A drive support device comprises parking row acquisition units 11, 12, 13A which acquire a parking frame PL of circumference of a self vehicle 100 and / or a parking vehicle VL and at the same time acquire a parking row PR on which the acquired parking frame PL and / or parking vehicle VL continue on a first threshold value or more, a parking area inside determination unit 16 which determines whether there is the self vehicle 100 within a parking area P which includes the parking row PR, an error operation determination unit 17 which acquires operation state of an acceleration operator and at the same time determines whether or not error operation of mistakingly stepping down the acceleration operator by a crew member has been executed on the basis of the operation state and a control unit 18 which executes driving force suppression control which restrains driving force of the self vehicle 100 when the parking area inside determination unit 16 determines that there is the self vehicle 100 within the parking area P and the error operation determination unit 17 determines that the error operation has been executed by the crew member.SELECTED DRAWING: Figure 14

Description

[Technical Field]

[0001] The present disclosure relates to a driving assistance device, a driving assistance method, and a program. [Background technology]

[0002] Conventionally, there is known a device that, when the amount of operation of a vehicle's accelerator pedal suddenly increases, determines that the driver has operated the accelerator pedal erroneously and executes driving force suppression control to suppress the vehicle's driving force (see, for example, Patent Documents 1 and 2, etc.). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-049981 [Patent Document 2] Patent No. 6299179 Summary of the Invention

[0004] There are situations in which a driver intentionally depresses the accelerator pedal heavily while a vehicle is traveling on a general road. Examples of such situations include when the vehicle is overtaking a preceding vehicle, when the vehicle is merging from an acceleration lane into a main lane, or when the vehicle is starting from an uphill road. In such cases, if it is determined that the driver has erroneously operated the accelerator pedal, the driving force suppression control will be unnecessarily activated even if the driver intends to accelerate. To prevent such unnecessarily activated driving force suppression control on a general road, it is possible to enable activation of the driving force suppression control when the vehicle is in a parking lot.

[0005] The device described in Patent Document 2 executes driving force suppression control when the vehicle is in a parking lot and the driver erroneously operates the accelerator pedal. However, the device described in Patent Document 2 determines that the vehicle is in a parking lot when it detects parking frame lines. Therefore, for example, when the device detects road markings such as stop lines at an intersection or pedestrian crossings while the vehicle is traveling on a public road, it may erroneously detect them as parking frame lines. In other words, there is a possibility that the device may erroneously determine that the vehicle is in a parking lot even when the vehicle is actually on a public road, and there is room for improvement in preventing unnecessary activation of the driving force suppression control.

[0006] The present disclosure has been made to solve the above-mentioned problems. That is, one of the objects of the present disclosure is to effectively prevent unnecessary operation of driving force suppression control.

[0007] The device of the present disclosure comprises: a parking row acquisition unit (11, 12, 13A) that detects parking spaces (PL) and / or parked vehicles (VL) around the vehicle (100) based on image data of the surroundings of the vehicle (100) and acquires a parking row (PR) in which the detected parking spaces (PL) and / or parked vehicles (VL) are adjacent to each other in a predetermined direction and are continuous for a predetermined first threshold number or more; a parking lot interior determination unit (16) that determines whether the vehicle (100) is present in a parking lot (P) having the parking row (PR); an erroneous operation determination unit (17) that acquires an operation state of an acceleration operator operated by an occupant of the vehicle (100) and determines whether the occupant has erroneously depressed the acceleration operator based on the operation state; and a control unit (18) that executes driving force suppression control to suppress the driving force of the vehicle (100) when the parking lot determination unit (16) determines that the vehicle (100) is present in the parking lot (P) and the erroneous operation determination unit (17) determines that the erroneous operation has been performed by the occupant.

[0008] The method of the present disclosure comprises: Based on image data of the surroundings of the vehicle (100), parking spaces (PL) and / or parked vehicles (VL) around the vehicle (100) are detected, and a parking row (PR) in which the detected parking spaces (PL) and / or parked vehicles (VL) are adjacent to each other in a predetermined direction and are continuous for a predetermined first threshold number or more is acquired; Determine whether the vehicle (100) is present in a parking lot (P) having the parking row (PR), acquiring an operation state of an acceleration operator operated by an occupant of the vehicle (100), and determining whether the occupant has erroneously depressed the acceleration operator based on the operation state; When it is determined that the vehicle (100) is present in the parking lot (P) and that the erroneous operation has been performed by the occupant, a driving force suppression control is executed to suppress the driving force of the vehicle (100).

[0009] The program of the present disclosure is A computer (10) of the driving assistance device (1) Based on image data of the surroundings of the vehicle (100), parking spaces (PL) and / or parked vehicles (VL) around the vehicle (100) are detected, and a parking row (PR) in which the detected parking spaces (PL) and / or parked vehicles (VL) are adjacent to each other in a predetermined direction and are continuous for a predetermined first threshold number or more is acquired; Determine whether the vehicle (100) is present in a parking lot (P) having the parking row (PR), acquiring an operation state of an acceleration operator operated by an occupant of the vehicle (100), and determining whether the occupant has erroneously depressed the acceleration operator based on the operation state; When it is determined that the vehicle (100) is present in the parking lot (P) and that the erroneous operation has been performed by the occupant, a process is executed to execute driving force suppression control to suppress the driving force of the vehicle (100).

[0010] According to the above configuration, when a predetermined number of adjacent parking spaces (PL) or parked vehicles (VL) are consecutively arranged in a predetermined direction and the number of adjacent parking spaces (PL) or parked vehicles (VL) is equal to or exceeds a predetermined first threshold number (for example, five), the parking spaces or parked vehicles are determined to be a parking row (PR). This effectively prevents erroneous determination of road markings such as stop lines or crosswalks painted on the surface of a general road, or other vehicles parked around the host vehicle (100) while waiting at a traffic light, as a parking row, and makes it possible to determine with high accuracy whether the host vehicle (100) is in a parking lot (P). Furthermore, by being able to determine with high accuracy whether the host vehicle (100) is in a parking lot (P), it also makes it possible to effectively prevent unnecessary activation of the driving force suppression control on a general road, etc.

[0011] In another aspect of the present disclosure, a travel locus prediction unit (15) that predicts a travel locus (TP) of the vehicle (100) based on the travel state of the vehicle (100); The parking lot interior determination unit (16) determines that the vehicle (100) is present in the parking lot (P) when the predicted travel path (TP) intersects with the parking row (PR).

[0012] According to this aspect, when the travel path (TP) of the host vehicle (100) intersects with the parking row (PR), in other words, when there is a high possibility that the host vehicle (100) will enter the parking row (PR), it is determined that the host vehicle (100) is in a parking lot (P), and activation of the driving force suppression control is enabled. This makes it possible to reliably improve safety while effectively preventing unnecessary activation of the driving force suppression control on general roads, etc.

[0013] In another aspect of the present disclosure, a parking row passage determination unit (14) that determines an area (E) between the plurality of parking rows (PR) as a parking row passage when the parking row acquisition unit (11, 12, 13A) acquires the plurality of parking rows (RP) that are opposed to each other across the vehicle (100) and the plurality of parking rows (PR) are in a predetermined positional relationship; The parking lot interior determination unit (16) determines that the vehicle (100) is located in the area (E) determined to be the inter-parking row passage when the vehicle (100) is located in the area (E) determined to be the inter-parking row passage.

[0014] According to this aspect, when a plurality of parking rows (PR) facing each other across the vehicle (100) are acquired, if the vehicle (100) is located within an area (E) of an aisle between these parking rows (PR), it is determined that the vehicle (100) is located in a parking lot (P). As a result, even if the predicted driving trajectory (TP) of the vehicle (100) does not intersect with the parking rows (PR) in a parking lot (P) such as a parking lot of a large store where parking rows (PR) are arranged on both sides of an aisle (R), for example, it becomes possible to effectively operate the driving force suppression control.

[0015] In another aspect of the present disclosure, a structure acquisition unit (13C) that acquires structures (ST) present around the vehicle (100) based on target information around the vehicle (100); a temporary parking row acquisition unit (13B) that detects parking spaces (PL) and / or parked vehicles (VL) around the vehicle (100) based on the image data, and acquires a temporary parking row (PRT) in which the detected parking spaces (PL) and / or parked vehicles (VL) are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being smaller than the first threshold number; When the structure acquisition unit (13C) acquires a structure (ST) of a predetermined length or more that is located on the opposite side of the temporary parking row (PRT) from the vehicle (100), and when the structure (ST) and the temporary parking row (PRT) have a predetermined positional relationship, the parking row acquisition unit (13A) acquires the temporary parking row (PRT) as a parking row (PR).

[0016] According to this aspect, when a temporary parking row (PRT) is acquired around the vehicle (100) in which the number of consecutive parking spaces (PL) or parked vehicles (VL) is equal to or greater than a second threshold number (e.g., 3) and less than a first threshold number (e.g., 5), and if a structure (ST) such as a building is present within a predetermined distance behind the temporary parking row (PRT), the temporary parking row (PRT) is determined to be a parking row (PR). This makes it possible to effectively operate the driving force suppression control even in a relatively small parking lot with a small number of parking spaces, such as a parking lot for a small store, and to reliably improve safety.

[0017] In another aspect of the present disclosure, a temporary parking row acquisition unit (13B) that detects parking spaces (PL) and / or parked vehicles (VL) around the vehicle (100) based on the image data, and acquires a temporary parking row (PRT) in which the detected parking spaces (PL) and / or parked vehicles (VL) are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being smaller than the first threshold number; When the parking row acquisition unit (13A) acquires at least one parking row (PR), if the parking row (PR) and the temporary parking row (PRT) are in a predetermined positional relationship, it acquires the temporary parking row (PRT) as the parking row (PR).

[0018] According to this aspect, when a parking row (PR) in which the number of consecutive parking spaces (PL) or parked vehicles (VL) is equal to or greater than a first threshold number (e.g., five) is acquired around the vehicle (100), and a temporary parking row (PRT) in which the number of consecutive spaces is equal to or greater than a second threshold number (e.g., three) but less than the first threshold number is further acquired, if the parking row (PR) and the temporary parking row (PRT) are in a predetermined positional relationship, the temporary parking row (PRT) is determined to be a parking row (PR). This makes it possible to effectively detect a small number of parking spaces (PL) or parked vehicles (VL), such as at the end of an aisle (R) in a parking lot (P) of a large store, as a parking row (PR). Furthermore, the driving force suppression control can be effectively activated even for a small number of parking spaces (PL) or parked vehicles (VL), thereby reliably improving safety.

[0019] In the above description, in order to facilitate understanding of the invention, the symbols used in the embodiments are added in parentheses to the constituent elements of the invention corresponding to the embodiments, but each constituent element of the invention is not limited to the embodiments defined by the symbols. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a schematic overall configuration diagram of a driving assistance device according to a first embodiment; [Figure 2] FIG. 1 is a schematic diagram illustrating an example of a parking lot demarcation line painted on the road surface. [Figure 3] FIG. 1 is a schematic diagram illustrating an example of a parking lot demarcation line painted on the road surface. [Figure 4] FIG. 1 is a schematic diagram illustrating an example of a parking lot demarcation line painted on the road surface. [Figure 5] FIG. 1 is a schematic diagram illustrating an example of a parking lot demarcation line painted on the road surface. [Figure 6] FIG. 1 is a schematic diagram illustrating an example of a parking lot demarcation line painted on the road surface. [Figure 7] FIG. 1 is a schematic diagram showing a top view of a vehicle parked in a parking lot. [Figure 8] FIG. 10 is a schematic diagram illustrating determination of a parking row based on adjacent parking spaces. [Figure 9] FIG. 10 is a schematic diagram illustrating determination of a parking row based on the contour lines of adjacent parked vehicles. [Figure 10] 10 is a schematic diagram illustrating determination of a parking row based on adjacent parking frames and parked vehicle contours; FIG. [Figure 11] FIG. 2 is a schematic diagram illustrating a rectangular frame defining a parking row. [Figure 12] FIG. 2 is a schematic diagram illustrating a rectangular frame defining a parking row. [Figure 13] FIG. 2 is a schematic diagram illustrating a rectangular frame defining a parking row. [Figure 14] FIG. 2 is a schematic top view showing a vehicle traveling through an aisle in a parking lot and a row of parking spaces. [Figure 15]4 is a flowchart illustrating a routine for processing parking row determination, parking lot interior determination, and driving force suppression control according to the first embodiment. [Figure 16] FIG. 10 is a schematic overall configuration diagram of a driving assistance device according to a second embodiment. [Figure 17] FIG. 1 is a schematic diagram showing a pair of parking rows facing each other with a vertical gap between them as viewed from above. [Figure 18] FIG. 10 is a schematic diagram illustrating a determination of an aisle between parking rows. [Figure 19] 10 is a flowchart illustrating a routine for processing parking row determination, parking lot interior determination, and driving force suppression control according to a second embodiment. [Figure 20] FIG. 10 is a schematic overall configuration diagram of a driving assistance device according to a third embodiment. [Figure 21] FIG. 10 is a schematic diagram illustrating determination of a temporary parking row. [Figure 22] FIG. 10 is a schematic diagram illustrating determination of a temporary parking row. [Figure 23] FIG. 10 is a schematic diagram illustrating determination of a temporary parking row. [Figure 24] FIG. 2 is a schematic diagram illustrating the positional relationship between a temporary parking row and a structure. [Figure 25] 10 is a flowchart illustrating a routine for processing parking row determination, parking lot interior determination, and driving force suppression control according to a third embodiment. [Figure 26] FIG. 10 is a schematic overall configuration diagram of a driving assistance device according to a fourth embodiment. [Figure 27] 10 is a schematic diagram illustrating the positional relationship between a rectangular frame that defines a parking row and a temporary rectangular frame that defines a temporary parking row. FIG. [Figure 28] 10 is a flowchart illustrating a routine for processing parking row determination, parking lot interior determination, and driving force suppression control according to a fourth embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, a driving assistance device, a driving assistance method, and a program according to this embodiment will be described with reference to the drawings. The same components are denoted by the same reference numerals, and their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

[0022] [Overall configuration] FIG. 1 is a schematic diagram of the overall configuration of a driving assistance device 1 according to this embodiment. The driving assistance device 1 is mounted on a vehicle 100. The vehicle 100 on which the driving assistance device 1 is mounted will also be referred to as "host vehicle" hereinafter to distinguish it from other vehicles. The driving assistance device 1 has an ECU 10. The ECU 10 includes a microcomputer as its main component. ECU stands for Electronic Control Unit. The microcomputer includes a CPU, ROM, RAM, an interface, etc., and the CPU realizes various functions by executing instructions (programs, routines) stored in the ROM.

[0023] The ECU 10 is a control device that serves as a central control unit for executing driving force suppression control that suppresses the driving force of the vehicle 100 when the vehicle 100 is in a parking lot and the driver erroneously operates (missteps on) the accelerator pedal (acceleration operator). For this reason, the ECU 10 is communicably connected to a driving device 20, a steering device 21, a braking device 22, a vehicle state acquisition device 30, a surroundings recognition device 40, etc.

[0024] The drive unit 20 generates a drive force to be transmitted to the drive wheels of the vehicle 100. Examples of the drive unit 20 include an electric motor and an engine. In this embodiment, the vehicle 100 may be any of a hybrid electric vehicle (HEV), a plug-in hybrid electric vehicle (PHEV), a fuel cell electric vehicle (FCEV), an electric vehicle (BEV), and an engine vehicle. The steering unit 21 is, for example, an electric power steering device, and applies a steering force to the wheels of the vehicle 100. The braking unit 22 is, for example, a disc-type brake device, and applies a braking force to the wheels of the vehicle 100.

[0025] The vehicle state acquisition device 30 is a group of sensors that acquire the state of the vehicle 100. Specifically, the vehicle state acquisition device 30 includes a vehicle speed sensor 31, an accelerator sensor 32, a brake sensor 33, a steering angle sensor 34, a turn signal switch 35, and the like.

[0026] The vehicle speed sensor 31 detects the traveling speed (vehicle speed V) of the vehicle 100 and transmits the detected vehicle speed V to the ECU 10. The vehicle speed sensor 31 may be a wheel speed sensor. The accelerator sensor 32 detects the amount of operation of an accelerator pedal (not shown) by the driver and transmits the detected amount of operation of the accelerator to the ECU 10. The brake sensor 33 detects the amount of operation of a brake pedal (not shown) by the driver and transmits the detected amount of operation of the brake to the ECU 10. The steering angle sensor 34 detects the steering angle of a steering wheel (or steering shaft) (not shown) and transmits the detected steering angle to the ECU 10. The turn signal switch 35 detects the operation of a turn signal lever (not shown) by the driver. When the driver operates the turn signal lever clockwise or counterclockwise from the neutral position, the turn signal switch 35 transmits an ON signal to the ECU 10 indicating that the turn signal lever is being operated.

[0027] The surroundings recognition device 40 is a type of sensor that acquires target information related to targets around the vehicle 100. Specifically, the surroundings recognition device 40 includes a camera sensor 41, a radar sensor 42, and the like. Examples of the target information include surrounding vehicles, surrounding buildings, intersections, traffic lights, signs, parking lot dividing lines, white lines on roads, stop lines, and temporary stop lines. The target information around the vehicle 100 acquired by the surroundings recognition device 40 is transmitted to the ECU 10.

[0028] The camera sensor 41 captures images of the surroundings of the vehicle 100 and processes the captured image data to obtain an image of the surroundings of the vehicle 100. The camera sensor 41 is, for example, a stereo camera or a monocular camera, and a digital camera having an imaging element such as a CMOS or a CCD can be used.

[0029] In this embodiment, the camera sensor 41 includes a front camera 41A, a rear camera 41B, a left side camera 41C, and a right side camera 41D. The front camera 41A captures an image of the surrounding area in front of the vehicle 100 and transmits the generated front image data to the ECU 10. The rear camera 41B captures an image of the surrounding area behind the vehicle 100 and transmits the generated rear image data to the ECU 10. The left side camera 41C captures an image of the surrounding area on the left side of the vehicle 100 and transmits the generated left side image data to the ECU 10. The right side camera 41D captures an image of the surrounding area on the right side of the vehicle 100 and transmits the generated right side image data to the ECU 10. Note that, hereinafter, the multiple cameras 41A to 41D will be referred to simply as "camera sensor 41." Furthermore, the front image data, rear image data, left side image data, and right side image data will be referred to simply as "image data."

[0030] The radar sensor 42 detects targets present in the area surrounding the vehicle 100. The radar sensor 42 includes a millimeter-wave radar and / or a lidar. The millimeter-wave radar emits millimeter-wave radio waves (millimeter waves) and receives millimeter waves reflected by targets present within the emission range (reflected waves). The millimeter-wave radar acquires the relative distance between the vehicle 100 and the target, the relative speed between the vehicle 100 and the target, and the like based on the phase difference between the transmitted millimeter waves and the received reflected waves, the attenuation level of the reflected waves, and the time between transmitting the millimeter waves and receiving the reflected waves. The lidar sequentially scans a pulsed laser beam with a wavelength shorter than that of the millimeter waves in multiple directions and receives the reflected light reflected by the target to acquire the shape of targets detected around the vehicle 100, the relative distance between the vehicle 100 and the target, the relative speed between the vehicle 100 and the target, and the like.

[0031] In this embodiment, the radar sensor 42 includes a front radar 42A, a rear radar 42B, a left side radar 42C, and a right side radar 42D. The front radar 42A detects targets present in a region ahead of the vehicle 100 and transmits information about the detected front targets to the ECU 10. The rear radar 42B detects targets present in a region behind the vehicle 100 and transmits information about the detected rear targets to the ECU 10. The left side radar 42C detects targets present in a region behind the vehicle 100 and transmits information about the detected left side targets to the ECU 10. The right side radar 42D detects targets present in a region to the right of the vehicle 100 and transmits information about the detected right side targets to the ECU 10. Note that, hereinafter, the multiple radars 42A to 42D will be simply referred to as "radar sensors 42."

[0032] [First embodiment] Next, the details of the ECU 10 according to the first embodiment will be described. Focusing on its functions, the ECU 10 has, as some functional elements, a parking space acquisition unit 11, a parked vehicle acquisition unit 12, a parking row determination unit 13A, a traveling trajectory prediction unit 15, a parking lot interior determination unit 16, an erroneous operation determination unit 17, and a driving force suppression control unit 18. These functional elements will be described as being included in the ECU 10, which is an integrated piece of hardware, but some of these elements may also be provided in an ECU separate from the ECU 10. Furthermore, all or some of the functional elements of the ECU 10 may also be provided in an information processing device in a facility (e.g., a management center) that can communicate with the host vehicle 100.

[0033] The parking space acquisition unit 11 acquires a parking space in a parking lot based on image data of the surroundings of the vehicle 100 captured by the camera sensor 41. Figures 2 to 6 are schematic diagrams illustrating examples of demarcation lines 200 drawn on the road surface of a parking lot P. In Figures 2 to 6, reference numeral 300 denotes a parked vehicle (other vehicle) parked in the parking lot P, and reference numeral R denotes a passage R along which a vehicle entering the parking lot P can travel.

[0034] The parking space acquisition unit 11 extracts a demarcation line 200 from the image data captured by the camera sensor 41 by performing image analysis processing such as edge extraction, pattern matching, and feature point extraction on the image data, and acquires a parking space PL (see dashed line) based on the extracted demarcation line 200. Here, the demarcation line 200 refers to a white line, a yellow line, or the like painted on the road surface of the parking lot P to demarcate a parking space PL for parking one vehicle. Whether the extracted demarcation line 200 demarcates a parking space PL can be determined, for example, by comparing the dimensions of the area defined by the extracted demarcation line 200 with the standard parking space dimensions (width, depth) of a general public parking lot.

[0035] In the following, among the boundaries that form the rectangular parking stall PL, the boundary line facing the aisle R will be referred to as the front boundary line PL1. Also, the boundary line that extends approximately parallel to the front boundary line PL1 and is farther away from the aisle R than the front boundary line PL1 is referred to as the rear boundary line PL2. Also, the boundary line that is perpendicular to each of the boundaries PL1 and PL2 and is located on the left side when viewed from the aisle R side will be referred to as the left boundary line PL3, and the boundary line that is located on the right side will be referred to as the right boundary line PL4.

[0036] 2, the demarcation line 200 is drawn on the road surface as a solid line in the shape of a substantially rectangular frame. In this case, the parking space acquisition unit 11 extracts, from a pair of demarcation lines 210, 220 extending substantially parallel to the extension direction of the aisle R, the demarcation line 210 located on the aisle R side as a front boundary line PL1, and the demarcation line 220 located farther from the aisle R than the demarcation line 210 as a rear boundary line PL2. In addition, from a pair of demarcation lines 230, 240 intersecting the demarcation lines 210, 220 at substantially right angles, the parking space acquisition unit 11 extracts, as a left-side boundary line PL3, the demarcation line 230 located on the left side as viewed from the aisle R side, and the demarcation line 240 located on the right side as a left-side boundary line PL4.

[0037] 3, the demarcation line 200 is drawn as two parallel straight lines extending in a direction substantially perpendicular to the extension direction of the aisle R, and a vehicle is parked between these two parallel straight lines. In this case, of the two parallel demarcation lines 230, 240 drawn on the road surface, the parking space acquisition unit 11 extracts the demarcation line 230 located on the left side as viewed from the aisle R as a left boundary line PL3 and the demarcation line 240 located on the right side as a right boundary line PL4. In addition, of the longitudinal ends of each demarcation line 230, 240, the parking space acquisition unit 11 extracts a first imaginary demarcation line 210S connecting the ends on the aisle R side as a front boundary line PL1 and a second imaginary demarcation line 220S connecting the ends on the opposite side from the aisle R as a rear boundary line PL2.

[0038] 4, the demarcation line 200 is drawn with two parallel double straight lines (double demarcation lines 230, 240) extending in a direction substantially perpendicular to the extension direction of the aisle R. Note that, hereinafter, of the respective demarcation lines 230A, 230B, 240A, 240B of the double demarcation lines 230, 240, the demarcation lines facing each other are referred to as the inner demarcation lines 230B, 240B. Of the inner demarcation lines 230B, 240B of the double demarcation lines 230, 240 drawn on the road surface, the parking space acquisition unit 11 extracts the inner demarcation line 230B located on the left side as viewed from the aisle R as the left boundary line PL3 and the inner demarcation line 240B located on the right side as the right boundary line PL4. In addition, the parking space acquisition unit 11 extracts the first virtual dividing line 210S connecting the ends on the aisle R side of the longitudinal ends of each inner dividing line 230B, 240B as the front boundary line PL1, and the second virtual dividing line 220S connecting the ends on the opposite side of the aisle R as the rear boundary line PL2.

[0039] 5, the demarcation line 200 is drawn as two straight lines that face each other in parallel at a predetermined distance and are inclined at a predetermined angle with respect to the extension direction of the aisle R. In this case, of the two demarcation lines 230, 240 drawn on the road surface, the parking space acquisition unit 11 extracts the demarcation line 230 located on the left side as viewed from the aisle R as the left boundary line PL3 and the demarcation line 240 located on the right side as the right boundary line PL4. In addition, the parking space acquisition unit 11 extracts the first imaginary demarcation line 210S extending from the end of the demarcation line 230 on the aisle R side toward the demarcation line 240 at a substantially right angle as the front boundary line PL1, and the second imaginary demarcation line 220S extending from the end of the demarcation line 240 opposite the aisle R toward the demarcation line 230 at a substantially right angle as the rear boundary line PL2.

[0040] 6 shows a state in which the parked vehicle 300 protrudes into the aisle R, making it impossible to extract a portion of the demarcation line 210. In this case, the parking space acquisition unit 11 extracts the demarcation line 230 that is not hidden by the parked vehicle 300 as the left boundary line PL3, the demarcation line 240 as the right boundary line PL4, and the demarcation line 220 as the rear boundary line PL2. In addition, the parking space acquisition unit 11 extracts the portions 210A and 210B of the demarcation line 210 that are not hidden by the parked vehicle 300 and the virtual demarcation line 210S that connects the ends of the unhidden portions 210A and 210B as the front boundary line PL1.

[0041] 6 illustrates an example in which the demarcation line 210 on the aisle R side is hidden, but similar processing is performed in cases in which the other demarcation lines 220, 230, and 240 are hidden by the parked vehicle 300, or in cases in which the demarcation lines 210-240 are hidden by obstacles other than the parked vehicle 300, or in cases in which some of the demarcation lines 210-240 have disappeared due to deterioration or the like, and therefore a description of these cases will be omitted. Also, if the demarcation line 200 cannot be extracted as a whole, the outline of the parked vehicle 300 acquired by the parked vehicle acquisition unit 12 may be used preferentially in determining the parking row, which will be described later.

[0042] The parking space acquisition unit 11 acquires position information of these extracted boundary lines PL1, PL2, PL3, and PL4 relative to the vehicle 100 (for example, coordinates in an xy plane coordinate system with the position of the vehicle 100 as the origin). Furthermore, the parking space acquisition unit 11 transmits the position information of the acquired boundary lines PL1, PL2, PL3, and PL4 to the parking row determination unit 13A at a predetermined cycle. Note that the types of demarcation lines 200 drawn on the road surface of the parking lot P are not limited to the examples shown in FIGS. 2 to 5, and may be other demarcation lines, such as when these demarcation lines 200 are mixed together or when they are drawn as dashed lines.

[0043] 1 again, the parked vehicle acquisition unit 12 acquires a vehicle contour line (hereinafter referred to as the parked vehicle contour line) that is the boundary between the parked vehicle 300 and the road surface based on image data of the area around the vehicle 100 captured by the camera sensor 41. FIG. 7 is a schematic diagram of a parked vehicle 300 parked in a parking lot as viewed from above. In FIG. 7, the symbol VL indicates the parked vehicle contour line. Note that the actual parked vehicle contour line VL has a complex shape that includes curves due to the protrusion of the side mirrors, the bumper, etc., but in this embodiment, the parked vehicle contour line VL will be described as the smallest rectangular frame line that can accommodate the outer periphery of the body of the parked vehicle 300.

[0044] The parked vehicle acquisition unit 12 first determines whether or not a parked vehicle 300 is captured in the image data by performing image analysis processes such as edge extraction, pattern matching, and feature point extraction on the image data captured by the camera sensor 41. Furthermore, if the parked vehicle acquisition unit 12 determines that a parked vehicle 300 is captured in the image data, it identifies the smallest rectangular frame line within the image data that can contain the parked vehicle 300, and extracts the identified rectangular frame line as the parked vehicle outline VL.

[0045] Specifically, the parked vehicle acquisition unit 12 extracts, from the identified frame line, a portion corresponding to the front end of the parked vehicle 300 as a front boundary line LV1. Furthermore, the parked vehicle acquisition unit 12 extracts, from the identified frame line, a portion corresponding to the rear end of the parked vehicle 300 as a rear boundary line LV2. Furthermore, the parked vehicle acquisition unit 12 extracts, from the identified frame line, a portion corresponding to the left end of the parked vehicle 300 as a left-side boundary line LV3. Furthermore, the parked vehicle acquisition unit 12 extracts, from the identified boundary line, a portion corresponding to the right end of the parked vehicle 300 as a right-side boundary line LV4. The parked vehicle acquisition unit 12 acquires position information of each of these extracted contour lines LV1 to LV4 relative to the host vehicle 100 (for example, coordinates in an xy plane coordinate system with the position of the host vehicle 100 as the origin), and transmits the acquired position information to the parking row determination unit 13A at a predetermined interval.

[0046] The parking row determination unit 13A determines whether the parking stalls PL and the parked vehicle contour lines VL form a continuous parking row based on the position information of the parking stalls PL transmitted from the parking space acquisition unit 12 and the position information of the parked vehicle contour lines VL transmitted from the parked vehicle acquisition unit 11B. Specific parking row determination processing will be described below with reference to FIGS. 8 to 12. In the following description, the longitudinal direction of the parking stalls PL and the parked vehicle contour lines VL is defined as the "longitudinal direction," and the direction substantially perpendicular to the longitudinal direction is defined as the "lateral direction." In the following description, the parking stalls PL and the parked vehicle contour lines VL will be described as an example of side-by-side parking where the parking stalls PL and the parked vehicle contour lines VL are adjacent in the lateral direction. However, the same processing is performed in the case of parallel parking where the parking stalls PL and the parked vehicle contour lines VL are adjacent in the longitudinal direction, and therefore, the description of parallel parking will be omitted.

[0047] FIG. 8 is a schematic diagram illustrating a determination based on adjacent parking stalls PL. When adjacent parking stalls PL are obtained from image data, the parking row determination unit 13A calculates a vertical separation distance DH1 between these front boundary lines PL1 based on the position information of their front boundary lines PL1. In addition, the parking row determination unit 13A determines whether the calculated separation distance DH1 satisfies a first condition that is equal to or less than a predetermined first threshold. Note that the first condition may be determined based on the separation distance between the rear boundary lines PL2 of adjacent parking stalls PL.

[0048] In addition, the parking row determination unit 13A calculates a lateral separation distance DH2 between the left and right boundary lines PL3, PL4 of adjacent parking stalls PL (the right boundary line PL4 of the parking stall PL on the left side in the figure and the left boundary line PL3 of the parking stall PL on the right side in the figure) based on the position information of these left and right boundary lines PL3, PL4. In addition, the parking row determination unit 13A determines whether the calculated separation distance DH2 satisfies a second condition that is equal to or less than a predetermined second threshold. The first threshold and the second threshold are not particularly limited, but may be set based on standard values ​​for general public parking lots. If both the first condition and the second condition are satisfied, the parking row determination unit 13A regards these adjacent parking stalls PL as being consecutive in the lateral direction.

[0049] FIG. 9 is a schematic diagram illustrating a determination based on adjacent parked vehicle contour lines VL. When adjacent parked vehicle contour lines VL are obtained from image data, the parking row determination unit 13A calculates the vertical separation distance DH3 between these front contour lines VL1 based on the position information of their front contour lines VL1. The parking row determination unit 13A also determines whether the calculated separation distance DH3 satisfies a third condition that the calculated separation distance DH3 is equal to or less than a predetermined third threshold. The third condition may also be determined based on the separation distance between the rear contour lines VL2 of adjacent parked vehicle contour lines VL.

[0050] Furthermore, the parking row determination unit 13A calculates a lateral separation distance DH4 between the left and right contour lines VL3, VL4 of adjacent parked vehicle contour lines VL (the right contour line VL4 of the parked vehicle contour line VL on the left side in the drawing and the left contour line PL3 of the parked vehicle contour line VL on the right side in the drawing) based on position information of these left and right contour lines VL3, VL4. The parking row determination unit 13A also determines whether the calculated separation distance DH4 satisfies a fourth condition, that is, whether the calculated separation distance DH4 is equal to or less than a predetermined fourth threshold. The third and fourth thresholds are not particularly limited, but it is preferable to set at least the fourth threshold to a value greater than the second threshold. If both the third and fourth conditions are satisfied, the parking row determination unit 13A considers these adjacent parked vehicle contour lines VL to be horizontally continuous.

[0051] In the example shown in Fig. 9, the parked vehicles 300 are all described as backward-parked vehicles that have been parked by backing up into the parking area, but the parked vehicles 300 may all be forward-parked vehicles that have been parked by moving forward into the parking area, or one parked vehicle 300 may be a backward-parked vehicle and the other a forward-parked vehicle. When one parked vehicle is a backward-parked vehicle and the other is a forward-parked vehicle, the third condition described above may be determined based on the distance between the front contour line VL1 of one parked vehicle contour line VL and the rear contour line VL2 of the other parked vehicle contour line VL. Whether the parked vehicle 300 is a forward-parked vehicle or a backward-parked vehicle may be determined by performing machine learning such as pattern matching on the image data.

[0052] FIG. 10 is a schematic diagram illustrating a determination based on adjacent parking stalls PL and parked vehicle contour lines VL. When the parking stalls PL and parked vehicle contour lines VL are obtained from image data, the parking row determination unit 13A calculates a vertical separation distance DH5 between the front boundary line PL1 and the front contour line VL1 based on the position information of the front boundary line PL1 of the parking stall PL and the position information of the front contour line VL1 of the parked vehicle contour line VL. In addition, the parking row determination unit 13A determines whether the calculated separation distance DH5 satisfies a fifth condition that is equal to or less than a predetermined fifth threshold. The fifth condition may be determined based on the separation distance between the rear boundary line PL2 of the parking stall PL and the rear contour line VL2 of the parked vehicle contour line VL.

[0053] Further, the parking row determination unit 13A calculates a lateral separation distance DH6 based on position information of the left and right boundary lines PL3, PL4 of the parking stall PL and the left and right contour lines VL3, VL4 of the parked vehicle contour line VL (in the illustrated example, the right boundary line PL4 of the parking stall PL on the left side in the drawing and the left contour line VL3 of the parked vehicle contour line VL on the right side in the drawing). Further, the parking row determination unit 13A determines whether the calculated separation distance DH6 satisfies a sixth condition that is equal to or less than a predetermined sixth threshold. The fifth threshold and the sixth threshold are not particularly limited, but it is preferable to set at least the sixth threshold to a value greater than the second threshold and smaller than the fourth threshold. When both the fifth condition and the sixth condition are satisfied, the parking row determination unit 13A regards the adjacent parking stall PL and parked vehicle contour line VL as being continuous in the lateral direction.

[0054] The parking row determination unit 13A determines that the number of consecutive parking stalls PL, the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL in no particular order is a parking row when the number is equal to or greater than a predetermined first threshold number (for example, 5). Specifically, as shown in FIGS. 11(A) and 11(B), when the number of consecutive parking stalls PL is equal to or greater than the first threshold number (5 in the illustrated example), the parking row determination unit 13A determines that the smallest rectangular frame PR into which the collection of parking stalls PL can fit is a parking row. Also, as shown in FIGS. 12(A) and 12(B), when the number of consecutive parked vehicle contour lines VL is equal to or greater than the first threshold number (5 in the illustrated example), the parking row determination unit 13A determines that the smallest rectangular frame PR into which the collection of parking stalls PL can fit is a parking row. Also, as shown in Figures 13(A) and 13(B), when the number of consecutive parking spaces PL and parked vehicle contour lines VL mixed together is equal to or greater than the first threshold number (in the illustrated example, a combination of three parking spaces PL and two parked vehicle contour lines VL), the parking row determination unit 13A determines that the smallest rectangular frame PR that can contain the collection of these is the parking row.

[0055] In this way, when the number of consecutive parking stalls PL, or consecutive parked vehicle contour lines VL, or consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than the first threshold number, the collection is determined to be a parking row, thereby effectively preventing the erroneous determination of road markings such as stop lines and crosswalks painted on the surface of a general road, or other vehicles parked around the vehicle 100 while waiting for a traffic light, etc., as a parking row.

[0056] The parking row determination unit 13A extracts a rectangular frame PR that defines the parking row from the image data, and acquires position information (e.g., coordinates in an xy plane coordinate system with the position of the vehicle 100 as the origin) of each of the straight lines PR1 to PR4 that form the extracted rectangular frame PR relative to the vehicle 100. The parking row determination unit 13A also transmits the acquired position information of each of the straight lines PR1 to PR4 to the parking lot interior determination unit 16 at a predetermined cycle. In the following, of the straight lines PR1 to PR4 that form the rectangular frame PR, the straight line PR1 facing the aisle R will be referred to as the front parking row line, and the straight line PR2 that is parallel to the front parking row line PR1 and is further away from the aisle R than the front parking row line PR1 is referred to as the rear parking row line. The straight line PR3 that is perpendicular to the front parking row line PR1 and the rear parking row line PR2 and is located on the left side as viewed from the aisle R will be referred to as the left parking row line, and the straight line PR4 that is located on the right side as viewed from the aisle R will be referred to as the right parking row line.

[0057] 1 again, the driving trajectory prediction unit 15 calculates a predicted driving trajectory of the host vehicle 100 based on the driving state of the host vehicle 100 acquired by the vehicle state acquisition device 30. Here, the predicted driving trajectory refers to a trajectory that the host vehicle 100 is predicted to travel if the current driving state of the host vehicle 100 is maintained. The predicted driving trajectory can be calculated based on, for example, the vehicle speed V acquired by the vehicle speed sensor 31 and the steering angle acquired by the steering angle sensor 34. The driving trajectory prediction unit 15 transmits the calculated predicted driving trajectory to the parking lot interior determination unit 16 at a predetermined cycle.

[0058] The parking lot interior determination unit 16 determines whether or not the host vehicle 100 is located within a parking lot P based on the position information of the parking row (rectangular frame PR) relative to the host vehicle 100 transmitted from the parking row determination unit 13A and the predicted driving trajectory of the host vehicle 100 transmitted from the driving trajectory prediction unit 15. FIG. 14 is a schematic top view showing the host vehicle 100 and the parking row traveling along an aisle R within the parking lot P. The parking lot interior determination unit 16 first determines whether or not the predicted driving trajectory TP of the host vehicle 100 intersects with the front parking row line PR1 of the rectangular frame PR that defines the parking row. If it is determined that the predicted driving trajectory TP intersects with the front parking row line PR1, the parking lot interior determination unit 16 calculates a predicted arrival time TA for the host vehicle 100 to arrive at an intersection position CP where the predicted driving trajectory TP and the front parking row line PR1 intersect from the current position. The predicted arrival time TA can be calculated, for example, by dividing the distance D along the predicted travel path TP from the current position of the vehicle 100 to the intersection position CP by the current vehicle speed V of the vehicle 100 (TA=D / V). The parking lot interior determination unit 16 determines that the vehicle 100 is present in the parking lot P if the calculated predicted arrival time TA is equal to or less than a predetermined threshold time (for example, several seconds).

[0059] When the parking lot interior determination unit 16 determines that the vehicle 100 is located in the parking lot P, the operation error determination unit 17 determines whether the driver of the vehicle 100 has erroneously depressed the accelerator pedal by mistake. Specifically, when all of the following conditions (1) to (5) are met, the operation error determination unit 17 determines that the driver has erroneously operated the accelerator pedal. On the other hand, when at least one of the conditions (1) to (5) is not met, the operation error determination unit 17 determines that the driver has not erroneously operated the accelerator pedal. (1) The vehicle speed V of the host vehicle 100 exceeds a predetermined vehicle speed threshold V Min is less than. (2) The accelerator pedal operation amount AP is equal to or exceeds a predetermined operation amount threshold value AP Max That's all. (3) The accelerator pedal operation speed APV is equal to or exceeds a predetermined operation speed threshold APV. Max That's all. (4) The brakes were not applied. (5) The turn signal is not operated.

[0060] The condition (1) may be determined based on the detection result of the vehicle speed sensor 31. Min is not particularly limited, and may be set based on a standard vehicle speed when the vehicle travels through aisle R in parking lot P, such as 10 to 15 km / h. Conditions (2) and (3) may be determined based on the detection results of accelerator sensor 32. The accelerator pedal operation speed APV of condition (3) may be determined by dividing the accelerator pedal operation amount AP acquired by accelerator sensor 32 by time. Condition (4) may be determined based on the detection results of brake sensor 33, and condition (5) may be determined based on the ON / OFF signal of turn signal switch 35. Note that the conditions for determining an erroneous operation of the accelerator pedal are not limited to the above conditions (1) to (5), and any of conditions (1) to (5) may be omitted, or other conditions may be added.

[0061] When the erroneous operation determination unit 17 determines that the driver has erroneously operated the accelerator pedal, the driving force suppression control unit 18 determines whether the actual acceleration GA of the host vehicle 100 is greater than or equal to a predetermined upper limit acceleration G Max The drive force suppression control is executed to control the operation of the drive unit 20 so that the upper limit acceleration G Max may be a constant value, or may be set to decrease as the vehicle speed V increases. After starting the driving force suppression control, the driving force suppression control unit 18 ends the driving force suppression control (i.e., the upper limit acceleration G Max In this way, if the driver erroneously operates the accelerator pedal in the parking lot P, the actual acceleration GA of the vehicle 100 is set to the upper limit acceleration G Max By executing drive force suppression control to keep the drive force below this level, safety can be improved.

[0062] Next, a processing routine for the parking line determination, parking lot in-car determination, and driving force suppression control by the ECU 10 will be described with reference to the flowchart shown in Fig. 15. When the ignition switch or start button of the vehicle 100 is turned ON, the ECU 10 repeatedly executes the processing from step S100 onwards in Fig. 15 at a predetermined cycle. Note that the ECU 10 may be configured to execute the processing from step S100 onwards when the vehicle speed V of the vehicle 100 is equal to or lower than a predetermined speed threshold.

[0063] In step S100, the ECU 10 searches for a parking space PL and a parked vehicle outline VL around the host vehicle 100 based on image data captured by the camera sensor 41.

[0064] Next, in step S105, the ECU 10 determines whether at least one of the parking space PL and the parked vehicle outline VL can be acquired from the image data. If at least one of the parking space PL and the parked vehicle outline VL can be acquired (Yes), the ECU 10 proceeds to the process of step S110. On the other hand, if neither the parking space PL nor the parked vehicle outline VL can be acquired (No), the ECU 10 returns the process to step S100 to continue searching for the parking space PL and the parked vehicle outline VL.

[0065] In step S110, the ECU 10 determines whether the vertical and horizontal separation distances between adjacent parking stalls PL and parked vehicle contour lines VL are equal to or less than predetermined thresholds. Specifically, when a plurality of adjacent parking stalls PL are acquired, the ECU 10 determines whether a first condition is satisfied that the vertical separation distance DH1 between these adjacent parking stalls PL is equal to or less than a first threshold, and a second condition is satisfied that the horizontal separation distance DH2 between these adjacent parking stalls PL is equal to or less than a second threshold. In addition, when adjacent parked vehicle contour lines VL are acquired, the ECU 10 determines whether a third condition is satisfied that the vertical separation distance DH3 between these adjacent parked vehicle contour lines VL is equal to or less than a third threshold, and a fourth condition is satisfied that the horizontal separation distance DH4 between these adjacent parked vehicle contour lines VL is equal to or less than a fourth threshold. In addition, when the adjacent parking stall PL and the parked vehicle contour line VL can be acquired, the ECU 10 judges whether or not a fifth condition is satisfied, that is, the vertical separation distance DH3 between the adjacent parking stall PL and the parked vehicle contour line VL is equal to or less than a fifth threshold, and a sixth condition is satisfied, that is, the horizontal separation distance DH4 between the adjacent parking stall PL and the parked vehicle contour line VL is equal to or less than a sixth threshold. If the condition is satisfied (Yes), the ECU 10 proceeds to the processing of step S112, judges that the adjacent parking stall PL and the parked vehicle contour line VL are continuous, and proceeds to the processing of step S115. On the other hand, if the condition is not satisfied in the judgment of step S110 (No), and if the adjacent parking stall PL and the parked vehicle contour line VL cannot be acquired, the ECU 10 returns the processing to step S100 to continue searching for the parking stall PL and the parked vehicle contour line VL.

[0066] In step S115, the ECU 10 determines whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a first threshold number. If the condition is met (Yes), the ECU 10 proceeds to the process of step S120, determines that they are a parking row, acquires position information of a rectangular frame PR that defines the parking row, and proceeds to step S50. On the other hand, if the condition is not met in the determination of step S115 (No), the ECU 10 returns the process to step S100 to continue searching for the parking stalls PL and the parked vehicle contour lines VL.

[0067] In step S150, the ECU 10 calculates a predicted traveling path TP of the vehicle 100 based on the vehicle speed V acquired by the vehicle speed sensor 31 and the steering angle acquired by the steering angle sensor 34. Next, in step S155, the ECU 10 determines whether the calculated predicted traveling path TP intersects with the front parking row line PR1 of the rectangular frame PR that defines the parking row. If they intersect (Yes), the ECU 10 proceeds to step S160. On the other hand, if they do not intersect (No), the ECU 10 returns the process to step S100 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0068] In step S160, the ECU 10 calculates a predicted arrival time TA that the vehicle 100 takes to reach the intersection position CP where the predicted travel path TP and the front parking row line PR1 intersect from the current position. Next, in step S165, the ECU 10 determines whether the predicted arrival time TA is equal to or shorter than a predetermined threshold time. If the predicted arrival time TA is equal to or shorter than the threshold time (Yes), the ECU 10 proceeds to step S170. On the other hand, if the predicted arrival time TA is not equal to or shorter than the threshold time (No), the ECU 10 returns the process to step S100 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0069] In step S170, the ECU 10 determines that the vehicle 100 is present in the parking lot P. Next, in step S175, the ECU 10 determines whether the driver has erroneously depressed the accelerator pedal. If all of the above conditions (1) to (5) are met (Yes), the ECU 10 determines that the driver has erroneously operated the accelerator pedal, and proceeds to step S180. On the other hand, if at least one of the above conditions (1) to (5) is not met (No), the ECU 10 returns the process to step S100 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0070] In step S180, the ECU 10 determines whether the actual acceleration GA of the host vehicle 100 is greater than or equal to a predetermined upper limit acceleration G MaxNext, in step S185, the ECU 10 determines whether the accelerator pedal operation amount AP has decreased to a predetermined end threshold APE or less. If the accelerator pedal operation amount AP has not decreased to a predetermined end threshold APE or less (No), the ECU 10 returns the process to step S180 to continue the driving force suppression control. On the other hand, if the accelerator pedal operation amount AP has decreased to a predetermined end threshold APE or less (Yes), the ECU 10 proceeds to step S190, ends the driving force suppression control, and then temporarily ends this routine. Thereafter, the ECU 10 repeatedly executes the processes of steps S100 to S190 described above until the ignition switch or start button of the vehicle 100 is turned OFF.

[0071] According to the first embodiment described above in detail, when the vertical and horizontal separation distances between adjacent parking stalls PL and parked vehicle contour lines VL are equal to or less than a predetermined threshold, they are determined to be consecutive. When the number of consecutive parking stalls PL, consecutive parked vehicle contour lines VL, or consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a first threshold number (for example, 5), the smallest rectangular frame PR into which they fit is determined to be a parking row. This effectively prevents road markings such as stop lines and crosswalks painted on the surface of a general road, or other vehicles parked around the host vehicle 100 while waiting for a traffic light, from being erroneously determined to be a parking row. Furthermore, by using the parking row determined in this way, it is possible to accurately determine whether the host vehicle 100 is in the parking lot P. Furthermore, if the predicted driving trajectory TP of the vehicle 100 intersects with the rectangular frame PR that defines the parking row, in other words, if there is a high possibility that the vehicle 100 will enter the parking row PR, it is determined that the vehicle 100 is in the parking lot P, and by enabling the operation of the driving force suppression control, it is possible to effectively prevent the driving force suppression control from operating unnecessarily on public roads, etc., while also reliably improving safety.

[0072] [Second embodiment] Next, a driving assistance device, a driving assistance method, and a program according to a second embodiment will be described. Fig. 16 is a schematic diagram of the overall configuration of a driving assistance device 1 according to the second embodiment. The driving assistance device 1 of the second embodiment is configured by adding a parking row passage determination unit 14 as a functional element to the ECU 10 of the first embodiment. The same components as those in the first embodiment are assigned the same reference numerals, and since their functions are also the same, detailed description thereof will be omitted.

[0073] When a pair of parking rows facing each other in the longitudinal direction with the vehicle 100 sandwiched between them is acquired around the vehicle 100, the parking row passage determination unit 14 determines whether or not the area between the rectangular frames PR defining these parking rows is a parking row passage. Fig. 17 is a schematic diagram showing a pair of parking rows facing each other with a gap between them in the longitudinal direction, viewed from above. In the example shown in Fig. 17, the front parking row lines PR1 of the rectangular frames PR defining the pair of parking rows are approximately the same length.

[0074] The parking row passage determination unit 14 first calculates the angle θ (preferably the angle formed by the front parking row line PR1) formed by the longitudinal straight lines PR1, PR2 among the straight lines PR1-PR4 of the rectangular frame PR that define the opposing parking rows. The parking row passage determination unit 14 also calculates the vertical separation distance between the opposing parking rows. Specifically, the parking row passage determination unit 14 calculates the lengths of the straight lines L1, L2 connecting both ends of the opposing front parking row line PR1, in other words, the straight lines L1, L2 connecting the opposing corners of each rectangular frame PR, as the separation distances DR1, DR2.

[0075] 18, if the difference in length between the front parking row line PR1 of one rectangular frame PR and the front parking row line PR1 of the other rectangular frame PR is large, the straight lines (see dashed lines) connecting the corners will be long. In such a case, the parking row passage determination unit 14 calculates the distances from the ends of the shorter front parking row line PR1 to the points where the straight lines L1 and L2 extending at approximately right angles thereto intersect with the longer front parking row line PR1 as the separation distances DR1 and DR2.

[0076] If the calculated angle θ is equal to or less than a predetermined threshold angle θV and the calculated separation distances DR1 and DR2 are both equal to or less than a predetermined distance threshold DRV, the parking row passage determination unit 14 determines that the area E enclosed by the opposing front parking row lines PR1 and the straight line connecting the ends of these opposing front parking row lines PR1 is an passage located between the parking rows (i.e., a parking row passage). Here, the threshold angle θV is not particularly limited, but may be set based on an angle (e.g., 10° or less) at which the opposing front parking row lines PR1 can be considered to be substantially parallel. The threshold angle θV may be a fixed value or may be a variable value depending on the length of the front parking row line PR1. The distance threshold DRV is also not particularly limited, but may be set based on a distance (e.g., approximately 8 to 10 m) that will not erroneously determine that the general road on which the vehicle 100 is traveling is an inter-parking row passage when the vehicle 100 detects a parking row across the general road while traveling on the general road.

[0077] When the parking row passage determination unit 14 determines that the space between the parking rows is an inter-parking row passage, and if the host vehicle 100 is located within area E, the parking lot interior determination unit 16 determines that the host vehicle 100 is located within parking lot P. In this way, by determining that the host vehicle 100 is located within parking lot P when the host vehicle 100 is located within area E between opposing parking rows, it becomes possible to effectively operate the driving force suppression control even in a parking lot where parking rows are provided on both sides of the aisle R, such as the parking lot of a large store, and the predicted driving trajectory TP described above does not intersect with the front parking row line PR1.

[0078] Next, a routine for processing parking row determination, parking lot interior determination, and driving force suppression control by the ECU 10 according to the second embodiment will be described based on the flowchart shown in FIG.

[0079] In step S200, the ECU 10 searches for the parking space PL and the parked vehicle outline VL around the vehicle 100 based on the image data captured by the camera sensor 41.

[0080] Next, in step S205, the ECU 10 determines whether at least one of the parking space PL and the parked vehicle outline VL can be acquired from the image data. If at least one of the parking space PL and the parked vehicle outline VL can be acquired (Yes), the ECU 10 proceeds to step S210. On the other hand, if neither the parking space PL nor the parked vehicle outline VL can be acquired (No), the ECU 10 returns to step S200 to continue searching for the parking space PL and the parked vehicle outline VL.

[0081] In step S210, the ECU 10 determines whether the vertical and horizontal separation distances between adjacent parking stalls PL and parked vehicle contour lines VL are equal to or less than a predetermined threshold. The specific processing is the same as that of the above-mentioned step S110 (see FIG. 15), so the description will be omitted. If the condition is met (Yes), the ECU 10 proceeds to the processing of step S212, determines that the adjacent parking stalls PL and parked vehicle contour lines VL are continuous, and proceeds to the processing of step S215. On the other hand, if the condition is not met in the determination of step S210 (No), the ECU 10 returns the processing to step S200 to continue searching for the parking stalls PL and parked vehicle contour lines VL.

[0082] In step S215, the ECU 10 determines whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a first threshold number. If the condition is met (Yes), the ECU 10 proceeds to the process of step S220, determines that they are a parking row, acquires position information of the rectangular frame PR that defines the parking row, and proceeds to step S225. On the other hand, if the condition is not met in the determination of step S215 (No), the ECU 10 returns the process to step S200 to continue searching for the parking stalls PL and the parked vehicle contour lines VL.

[0083] In step S225, the ECU 10 determines whether or not a plurality of opposing parking rows (rectangular frames PR) have been acquired. If a plurality of rectangular frames PR have been acquired (Yes), the ECU 10 proceeds to the processing of step S230. In step S230, the ECU 10 calculates the angle θ formed by the longitudinal straight lines (front parking row lines PR1) of the rectangular frames PR that define the opposing parking rows. Next, in step S232, the ECU 10 calculates the separation distances DR1, DR2 (the lengths of the straight lines L1, L2 connecting the ends of the front parking row lines PR1) between the rectangular frames PR that define the opposing parking rows. Note that the processing of steps S230 and S232 can be performed in any order.

[0084] In step S234, it is determined whether or not two conditions are met: the angle θ calculated in step S230 is equal to or less than a predetermined threshold angle θV, and both of the separation distances DR1 and DR2 calculated in step S232 are equal to or less than a predetermined distance threshold DRV. If at least one of the two conditions is not met (No), the ECU 10 proceeds to step S250, which will be described later. On the other hand, if both of the two conditions are met (Yes), the ECU 10 proceeds to step S236, where it determines that the area E surrounded by the opposing front parking row lines PR1 and the straight lines L1 and L2 connecting the ends of these opposing front parking row lines PR1 is an inter-parking row passage. Next, the ECU 10 proceeds to step S238.

[0085] In step S238, the ECU 10 determines whether the vehicle 100 is located within the area E. If the vehicle 100 is located within the area E (Yes), the ECU 10 proceeds to step S270, which will be described later. On the other hand, if the vehicle 100 is not located within the area E (No), the ECU 10 returns the process to step S200 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0086] In step S225, when it is determined that a plurality of rectangular frames PR have not been acquired (No), that is, when only one rectangular frame PR has been acquired, the ECU 10 proceeds to the process of step S250. In step S250, the ECU 10 calculates a predicted traveling trajectory TP of the host vehicle 100 based on the vehicle speed V acquired by the vehicle speed sensor 31 and the steering angle acquired by the steering angle sensor 34. Next, in step S255, the ECU 10 determines whether the calculated predicted traveling trajectory TP intersects with the front parking row line PR1 of the rectangular frame PR that defines the parking row. If they intersect (Yes), the ECU 10 proceeds to the process of step S260. On the other hand, if they do not intersect (No), the ECU 10 returns the process to step S200 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0087] In step S260, the ECU 10 calculates a predicted arrival time TA that the vehicle 100 takes to travel from the current position to the intersection position CP where the predicted travel path TP and the front parking row line PR1 intersect. Next, in step S265, the ECU 10 determines whether the predicted arrival time TA is equal to or shorter than a predetermined threshold time. If the predicted arrival time TA is equal to or shorter than the threshold time (Yes), the ECU 10 proceeds to step S270. On the other hand, if the predicted arrival time TA is not equal to or shorter than the threshold time (No), the ECU 10 returns the process to step S200 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0088] In step S270, the ECU 10 determines that the vehicle 100 is present in a parking lot. Next, in step S275, the ECU 10 determines whether the driver has erroneously depressed the accelerator pedal. If all of the above conditions (1) to (5) are met (Yes), the ECU 10 determines that the driver has erroneously operated the accelerator pedal, and proceeds to the process of step S280. On the other hand, if at least one of the above conditions (1) to (5) is not met (No), the ECU 10 returns the process to step S200 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0089] In step S280, the ECU 10 determines whether the actual acceleration GA of the host vehicle 100 is greater than or equal to a predetermined upper limit acceleration G Max Next, in step S285, the ECU 10 determines whether the accelerator pedal operation amount AP has decreased to a predetermined termination threshold APE or less. If the accelerator pedal operation amount AP has not decreased to a predetermined termination threshold APE or less (No), the ECU 10 returns the process to step S280 to continue the driving force suppression control. On the other hand, if the accelerator pedal operation amount AP has decreased to a predetermined termination threshold APE or less (Yes), the ECU 10 proceeds to step S290 to terminate the driving force suppression control, and then temporarily ends this routine. Thereafter, the ECU 10 repeatedly executes the processes of steps S200 to S290 described above until the ignition switch or start button of the vehicle 100 is turned OFF.

[0090] According to the second embodiment described above in detail, when multiple parking rows facing each other across the aisle R on both sides of the host vehicle 100 are acquired, and these parking rows are in a predetermined positional relationship, the area E between these parking rows is determined to be an inter-parking row aisle. Furthermore, when the host vehicle 100 is located within area E, the host vehicle 100 is determined to be located in the parking lot P regardless of whether the predicted driving trajectory TP intersects with the parking rows. As a result, even in a parking lot where parking rows are located on both sides of the aisle R, such as the parking lot of a large store, and the predicted driving trajectory TP of the host vehicle 100 does not intersect with the parking rows, it is possible to accurately determine that the host vehicle 100 is located in the parking lot P, and the driving force suppression control can be effectively operated within the parking lot P, thereby reliably improving safety.

[0091] [Third embodiment] Next, a driving assistance device, a driving assistance method, and a program according to a third embodiment will be described. FIG. 20 is a schematic overall configuration diagram of a driving assistance device 1 according to the third embodiment. The driving assistance device 1 of the third embodiment is configured by adding a temporary parking row determination unit 13B and a structure acquisition unit 13C as functional elements to the ECU 10 of the first embodiment. The same components as those in the first embodiment are assigned the same reference numerals, and since their functions are also the same, detailed description thereof will be omitted.

[0092] Even if the number of consecutive parking stalls PL, the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL in random order is less than the first threshold number (e.g., 5), if the number of consecutive parking stalls PL and parked vehicle contour lines VL is a predetermined second threshold number (e.g., 3 or more), the temporary parking row determination unit 13B determines them as a temporary parking row. Specifically, as shown in FIG. 21, if the number of consecutive parking stalls PL is less than the first threshold number and is equal to or greater than the second threshold number (3 in the illustrated example), the temporary parking row determination unit 13B determines the smallest temporary rectangular frame PRT into which the collection of parking stalls PL can fit as a temporary parking row. Also, as shown in FIG. 22, if the number of consecutive parked vehicle contour lines VL is less than the first threshold number and is equal to or greater than the second threshold number (3 in the illustrated example), the temporary parking row determination unit 13B determines the smallest temporary rectangular frame PRT into which the collection of parking stalls PL can fit as a temporary parking row. Also, as shown in FIG. 23, when the number of consecutive parking spaces PL and parked vehicle contour lines VL is less than the first threshold number and is equal to or greater than the second threshold number (in the illustrated example, a combination of two parking spaces PL and one parked vehicle contour line VL), the temporary parking row determination unit 13B determines that the smallest temporary rectangular frame PRT that can contain the collection of these is the temporary parking row.

[0093] The temporary parking row determination unit 13B extracts a temporary rectangular frame PRT that defines the temporary parking row from the image data, and acquires position information (for example, coordinates in an xy plane coordinate system with the position of the vehicle 100 as the origin) of each of the straight lines PRT1 to PRT4 that form the extracted temporary rectangular frame PRT relative to the vehicle 100. The temporary parking row determination unit 13B also transmits the acquired position information of each of the straight lines PRT1 to PRT4 to the parking row determination unit 13A at a predetermined cycle. In the following, of the straight lines PRT1 to PRT4 that form the temporary rectangular frame PRT, the straight line PRT1 facing the aisle R will be referred to as the front temporary parking row line, and the straight line PRT2 that is parallel to the front temporary parking row line PRT1 and is farther away from the aisle R than the front temporary parking row line PRT1 will be referred to as the rear temporary parking row line. In addition, the straight line PRT3 that is perpendicular to the front temporary parking line PRT1 and the rear temporary parking line PRT2 and is on the left side when viewed from the aisle R side is called the left temporary parking line, and the straight line PRT4 that is on the right side when viewed from the aisle R side is called the right temporary parking line.

[0094] The structure acquisition unit 13C acquires position information of a structure, such as a building, present in the rear area of ​​the temporary rectangular frame PR based on the detection result of the surroundings recognition device 40. Specifically, as shown in FIG. 24, the structure acquisition unit 13C first determines whether a structure ST having a predetermined length LV or more is present in the rear area of ​​the temporary rectangular frame PRT based on the detection result of the surroundings recognition device 40. Furthermore, if the structure acquisition unit 13C determines that a structure ST is present, it acquires a separation distance DB between the structure ST and the rear temporary parking row line PRT2 of the temporary rectangular frame RPT. Here, the rear area of ​​the temporary rectangular frame PR refers to the area on the opposite side of the temporary rectangular frame PR from the vehicle 100 (or the passage R). Furthermore, the structure ST is a concept that refers to a stationary object, such as a building erected from the ground, and does not include moving objects, such as other vehicles. Whether the structure ST is a building may be determined by performing machine learning, such as pattern matching, based on image data captured by the camera sensor 41. Furthermore, the predetermined length LV is not particularly limited, but it is desirable that it be at least longer than the rear temporary parking row line PRT2. The structure acquisition unit 13C transmits the acquired separation distance DB between the structure ST and the rear temporary parking row line PRT2 to the parking row determination unit 13A at a predetermined cycle.

[0095] The parking row determination unit 13A determines whether a temporary parking row can be considered a parking row based on the separation distance DB between the structure ST and the rear temporary parking row line PRT2 transmitted from the structure acquisition unit 13C. Specifically, if the separation distance DB between the structure ST and the rear temporary parking row line PRT2 is equal to or less than a predetermined seventh threshold, the parking row determination unit 13A determines that the temporary parking row defined by the temporary rectangular frame PRT is a parking row. The seventh threshold is not particularly limited, but may be set based on the standard distance between a small store building and parking spaces (e.g., 3 to 5 meters). In this way, even if the number of consecutive temporary parking rows is less than the first threshold number, if a structure ST is present in the area behind it, by determining the temporary parking row as a parking row, parking rows can be effectively detected, for example, even in parking lots with a small number of parking spaces, such as those of small stores.

[0096] The parking lot interior determination unit 16 determines whether or not the host vehicle 100 is located within a parking lot P based on the position information of the parking row (temporary rectangular frame PRT) relative to the host vehicle 100 transmitted from the parking row determination unit 13A and the predicted driving trajectory TP of the host vehicle 100 transmitted from the driving trajectory prediction unit 15. The parking lot interior determination unit 16 determines whether or not the predicted driving trajectory TP of the host vehicle 100 intersects with the front temporary parking row line PRT1 of the temporary rectangular frame PRT that defines the temporary parking row determined to be the parking row. If these lines intersect, the parking lot interior determination unit 16 calculates a predicted arrival time TA for the host vehicle 100 to reach the intersecting position from the current position, and if the calculated predicted arrival time TA is equal to or less than a predetermined threshold time, determines that the host vehicle 100 is located within the parking lot P.

[0097] Next, a routine for processing parking row determination, parking lot interior determination, and driving force suppression control by the ECU 10 according to the third embodiment will be described based on the flowchart shown in FIG.

[0098] In step S300, the ECU 10 searches for a parking stall PL and a parked vehicle outline VL around the vehicle 100 based on the image data captured by the camera sensor 41. Next, in step S305, the ECU 10 determines whether at least one of the parking stall PL and the parked vehicle outline VL has been acquired from the image data. If at least one of the parking stall PL and the parked vehicle outline VL has been acquired (Yes), the ECU 10 proceeds to step S310. On the other hand, if neither the parking stall PL nor the parked vehicle outline VL has been acquired (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0099] In step S310, the ECU 10 determines whether the vertical and horizontal separation distances between adjacent parking stalls PL and parked vehicle contour lines VL are equal to or less than a predetermined threshold. The specific processing is the same as that of the above-mentioned step S110 (see FIG. 15), so the description will be omitted. If the condition is met (Yes), the ECU 10 proceeds to the processing of step S312, determines that the adjacent parking stalls PL and parked vehicle contour lines VL are continuous, and proceeds to the processing of step S315. On the other hand, if the condition is not met in the determination of step S310 (No), the ECU 10 returns the processing to step S300 to continue searching for the parking stalls PL and parked vehicle contour lines VL.

[0100] In step S315, the ECU 10 determines whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a first threshold number. If the condition is met (Yes), the ECU 10 proceeds to the process of step S320, determines that they are a parking row, acquires position information of a rectangular frame PR that defines the parking row, and proceeds to step S350. On the other hand, if the condition is not met in the determination of step S315 (No), that is, if the consecutive number is less than the first threshold number, the ECU 10 proceeds to the process of step S330.

[0101] In step S330, the ECU 10 determines whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a second threshold number. If the condition is met (Yes), the ECU 10 proceeds to the process of step S332, determines that they are a temporary parking row, acquires position information of the temporary rectangular frame PRT that defines the temporary parking row, and proceeds to step S334. On the other hand, if the condition is not met in the determination of step S330 (No), that is, if the consecutive number is less than the second threshold number, the ECU 10 returns the process to step S300 to continue searching for the parking stalls PL and the parked vehicle contour lines VL.

[0102] In step S334, the ECU 10 determines whether a structure ST exists in the rear area of ​​the temporary rectangular frame PRT. If a structure ST exists (Yes), the ECU 10 proceeds to step S336. On the other hand, if a structure ST does not exist (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0103] In step S336, the ECU 10 determines whether the separation distance DB between the structure ST and the rear temporary parking row line PRT2 of the temporary rectangular frame PRT obtained in step S334 is equal to or less than a seventh threshold. If the separation distance DB is equal to or less than the seventh threshold (Yes), the ECU 10 proceeds to step S338, determines that the temporary parking row is a parking row, and proceeds to step S350. On the other hand, if it is determined in step S336 that the separation distance DB is less than the seventh threshold (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0104] In step S350, the ECU 10 calculates a predicted traveling path TP of the vehicle 100 based on the vehicle speed V acquired by the vehicle speed sensor 31 and the steering angle acquired by the steering angle sensor 34. Next, in step S355, the ECU 10 determines whether the calculated predicted traveling path TP intersects with the front parking row line PR1 of the rectangular frame PR or the front temporary parking row line PRT1 of the temporary rectangular frame PRT regarded as the parking row. If they intersect (Yes), the ECU 10 proceeds to step S360. On the other hand, if they do not intersect (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0105] In step S360, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach the rectangular frame PR or the temporary rectangular frame PRT from the current position. Specifically, if the predicted traveling trajectory TP intersects with the front parking row line PR1 of the rectangular frame PR, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach a position where the measured traveling trajectory TP and the front parking row line PR1 intersect from the current position. Also, if the predicted traveling trajectory TP intersects with the front temporary parking row line PRT1 of the temporary rectangular frame PRT, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach a position where the measured traveling trajectory TP and the front temporary parking row line PRT1 intersect from the current position.

[0106] Next, in step S365, the ECU 10 determines whether the predicted arrival time TA is equal to or shorter than a predetermined threshold time. If the predicted arrival time TA is equal to or shorter than the threshold time (Yes), the ECU 10 proceeds to step S370. On the other hand, if the predicted arrival time TA is not equal to or shorter than the threshold time (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0107] In step S370, the ECU 10 determines that the vehicle 100 is present in the parking lot P. Next, in step S375, the ECU 10 determines whether the driver has erroneously depressed the accelerator pedal. If all of the above conditions (1) to (5) are met (Yes), the ECU 10 determines that the driver has erroneously depressed the accelerator pedal, and proceeds to step S380. On the other hand, if at least one of the above conditions (1) to (5) is not met (No), the ECU 10 returns the process to step S300 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0108] In step S380, the ECU 10 determines whether the actual acceleration GA of the host vehicle 100 is greater than or equal to a predetermined upper limit acceleration G Max Next, in step S385, ECU 10 determines whether accelerator pedal depression amount AP has decreased to or below a predetermined termination threshold APE. If accelerator pedal depression amount AP has not decreased to or below termination threshold APE (No), ECU 10 returns the process to step S380 to continue the driving force suppression control. On the other hand, if accelerator pedal depression amount AP has decreased to or below termination threshold APE (Yes), ECU 10 proceeds to step S390, terminates the driving force suppression control, and then temporarily ends this routine. Thereafter, ECU 10 repeatedly executes the processes of steps S300 to S390 described above until the ignition switch or start button of vehicle 100 is turned OFF.

[0109] According to the third embodiment described above in detail, when a temporary parking row is acquired around the vehicle 100 in which the number of consecutive parking stalls PL or parked vehicle contour lines VL is equal to or greater than the second threshold number (e.g., 3) and less than the first threshold number (e.g., 5), and when a structure ST of a predetermined length or greater is present within a predetermined distance behind the temporary parking row, the temporary parking row is determined to be a parking row. This makes it possible to effectively detect parking rows even in relatively small parking lots with a small number of parking stalls, such as small stores. Furthermore, it becomes possible to effectively operate the driving force suppression control even in relatively small parking lots with a small number of parking stalls, thereby reliably improving safety.

[0110] [Fourth embodiment] Next, a driving assistance device, a driving assistance method, and a program according to a fourth embodiment will be described. FIG. 26 is a schematic overall configuration diagram of a driving assistance device 1 according to the fourth embodiment. The driving assistance device 1 of the fourth embodiment is configured by adding a temporary parking row determination unit 13B as a functional element to the ECU 10 of the first embodiment. The same components as those in the first embodiment are assigned the same reference numerals, and since their functions are also the same, detailed description thereof will be omitted.

[0111] Even if the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL in no particular order is less than the first threshold number (for example, 5), if the number of consecutive parking stalls PL and parked vehicle contour lines VL is a predetermined second threshold number (for example, 3 or more), the temporary parking row determination unit 13B determines that they are a temporary parking row. Note that the specific processing content is the same as that of the third embodiment, and therefore detailed description will be omitted.

[0112] As in the first embodiment, when the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL in no particular order is equal to or greater than a first threshold number (for example, 5), the parking row determination unit 13A determines that they are a parking row, and acquires position information of the smallest rectangular frame PR that can fit into the collection of those, relative to the host vehicle 100. Furthermore, the parking row determination unit 13A determines whether the temporary parking row can be regarded as a parking row, based on the position information of the rectangular frame PR acquired as the parking row and the position information of the temporary rectangular frame PRT transmitted from the temporary parking row determination unit 13B.

[0113] Specifically, as shown in FIG. 27, if the shortest separation distance DS between the rectangular frame PR and the temporary rectangular frame PRT (in the illustrated example, the distance between the corners facing the aisle R) is equal to or less than a predetermined eighth threshold, the parking row determination unit 13A determines that the temporary parking row defined by the temporary rectangular frame PRT is a parking row. The eighth threshold is not particularly limited, but may be set based on the standard aisle width (e.g., 8 to 10 m) of a general public parking lot. In this way, even if the number of consecutive temporary parking stalls is less than the first threshold, if the temporary parking row is located close to a parking row with a number of consecutive stalls equal to or greater than the first threshold, the temporary parking row can be regarded as a parking row. For example, in a parking lot of a large store or the like, a small number of parking stalls PL located at the end of an aisle R can be effectively detected as a parking row.

[0114] The parking lot interior determination unit 16 determines whether the host vehicle 100 is located within a parking lot based on the position information of the parking row (rectangular frame PR, PRT) relative to the host vehicle 100 transmitted from the parking row determination unit 13A and the predicted driving trajectory TP of the host vehicle 100 transmitted from the driving trajectory prediction unit 15. As in the first embodiment, the parking lot interior determination unit 16 determines whether the predicted driving trajectory TP of the host vehicle 100 intersects with the front parking row line PR1 of the rectangular frame PR that defines the parking row. The parking lot interior determination unit 16 also determines whether the predicted driving trajectory TP of the host vehicle 100 intersects with the front temporary parking row line PRT1 of the temporary rectangular frame PRT that defines the temporary parking row regarded as the parking row. If it is determined that the predicted driving trajectory TP intersects with at least one of the parking row lines PR1, PRT1, the parking lot interior determination unit 16 calculates a predicted arrival time TA for the host vehicle 100 to reach the intersecting position from the current position. Furthermore, the parking lot interior determination unit 16 determines that the vehicle 100 is present in the parking lot P when the calculated predicted arrival time TA is equal to or shorter than a predetermined threshold time.

[0115] Next, a processing routine for the parking row determination, the in-parking lot determination, and the driving force suppression control by the ECU 10 according to the fourth embodiment will be described based on the flowchart shown in FIG.

[0116] In step S400, the ECU 10 searches for a parking space PL and a parked vehicle contour line VL around the vehicle 100 based on the image data captured by the camera sensor 41. Next, in step S405, the ECU 10 determines whether at least one of the parking space PL and the parked vehicle contour line VL has been acquired from the image data. If at least one of the parking space PL and the parked vehicle contour line VL has been acquired (Yes), the ECU 10 proceeds to step S410. On the other hand, if neither the parking space PL nor the parked vehicle contour line VL has been acquired (No), the ECU 10 returns the process to step S400 to continue searching for the parking space PL and the parked vehicle contour line VL.

[0117] In step S410, the ECU 10 determines whether the vertical and horizontal separation distances between adjacent parking stalls PL and parked vehicle contour lines VL are equal to or less than a predetermined threshold. The specific processing is the same as that of the above-mentioned step S110 (see FIG. 15), so the description will be omitted. If the condition is met (Yes), the ECU 10 proceeds to the processing of step S412, determines that the adjacent parking stalls PL and parked vehicle contour lines VL are continuous, and proceeds to the processing of step S415. On the other hand, if the condition is not met in the determination of step S410 (No), the ECU 10 returns the processing to step S400 to continue searching for the parking stalls PL and parked vehicle contour lines VL.

[0118] In step S415, the ECU 10 determines whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than a first threshold number. If the condition is met (Yes), the ECU 10 proceeds to the process of step S420, determines that they are a parking row, acquires position information of a rectangular frame PR defining the parking row, and proceeds to step S425. On the other hand, if the condition is not met in the determination of step S415 (No), that is, if the consecutive number is less than the first threshold number, the ECU 10 returns the process to step S400 to continue searching for the parking stalls PL and the parked vehicle contour lines VL. Note that, if the determination of step S415 shows that there are multiple continuums that satisfy the condition, the ECU 10 may determine each of the multiple continuums as a parking row in step S420.

[0119] In step S425, the ECU 10 determines whether or not there are other consecutive parking stalls PL, or parked vehicle contour lines VL, or a combination of the parking stalls PL and the parked vehicle contour lines VL (continuum) around the rectangular frame PR that defines the parking row acquired in step S420. If there are other continuums around the rectangular frame PR (Yes), the ECU 10 proceeds to step S430. On the other hand, if there are no other continuums around the rectangular frame PR (No), the ECU 10 proceeds to step S450.

[0120] In step S430, it is determined whether the number of consecutive parking stalls PL, or the number of consecutive parked vehicle contour lines VL, or the number of consecutive parking stalls PL and parked vehicle contour lines VL is equal to or greater than the second threshold number and less than the first threshold number. If the condition is met (Yes), the ECU 10 proceeds to the processing of step S432, determines that they are a temporary parking row, and acquires the position information of the temporary rectangular frame PRT that defines the temporary parking row.

[0121] Next, in step S434, the ECU 10 determines whether the shortest separation distance DS between the rectangular frame PR and the temporary rectangular frame PRT is equal to or less than a predetermined eighth threshold based on the position information of the rectangular frame PR acquired in step S420 and the position information of the temporary rectangular frame PRT acquired in step S432. Here, if a plurality of parking rows are acquired in step S420, the ECU 10 may calculate the shortest separation distance DS using the parking row closest to the temporary parking row among the plurality of parking rows. If the shortest separation distance DS is equal to or less than the eighth threshold (Yes), the ECU 10 proceeds to the processing of step S436, determines that the temporary parking row defined by the temporary rectangular frame PRT is a parking row, and proceeds to the processing of step S450. On the other hand, if it is determined in step S434 that the shortest separation distance DS is not equal to or less than the eighth threshold (No), the ECU 10 returns to the processing of step S400 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0122] In step S450, the ECU 10 calculates a predicted traveling path TP of the vehicle 100 based on the vehicle speed V acquired by the vehicle speed sensor 31 and the steering angle acquired by the steering angle sensor 34. Next, in step S455, the ECU 10 determines whether the calculated predicted traveling path TP intersects with the front parking row line PR1 of the rectangular frame PR or the front temporary parking row line PRT1 of the temporary rectangular frame PRT regarded as the parking row. If they intersect (Yes), the ECU 10 proceeds to step S460. On the other hand, if they do not intersect (No), the ECU 10 returns the process to step S400 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0123] In step S460, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach the rectangular frame PR or the temporary rectangular frame PRT from the current position. Specifically, if the predicted traveling trajectory TP intersects with the front parking row line PR1 of the rectangular frame PR, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach a position where the measured traveling trajectory TP and the front parking row line PR1 intersect from the current position. Also, if the predicted traveling trajectory TP intersects with the front temporary parking row line PRT1 of the temporary rectangular frame PRT, the ECU 10 calculates a predicted arrival time TA for the host vehicle 100 to reach a position where the measured traveling trajectory TP and the front temporary parking row line PR1 intersect from the current position.

[0124] Next, in step S465, the ECU 10 determines whether the predicted arrival time TA is equal to or shorter than a predetermined threshold time. If the predicted arrival time TA is equal to or shorter than the threshold time (Yes), the ECU 10 proceeds to step S470. On the other hand, if the predicted arrival time TA is not equal to or shorter than the threshold time (No), the ECU 10 returns the process to step S400 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0125] In step S470, the ECU 10 determines that the vehicle 100 is present in a parking lot. Next, in step S475, the ECU 10 determines whether the driver has erroneously depressed the accelerator pedal. If all of the above conditions (1) to (5) are met (Yes), the ECU 10 determines that the driver has erroneously operated the accelerator pedal, and proceeds to step S480. On the other hand, if at least one of the above conditions (1) to (5) is not met (No), the ECU 10 returns the process to step S400 to continue searching for the parking stall PL and the parked vehicle outline VL.

[0126] In step S480, the ECU 10 determines whether the actual acceleration GA of the host vehicle 100 is greater than or equal to a predetermined upper limit acceleration G MaxNext, in step S485, ECU 10 determines whether accelerator pedal operation amount AP has decreased to or below a predetermined termination threshold APE. If accelerator pedal operation amount AP has not decreased to or below the termination threshold APE (No), ECU 10 returns the process to step S480 to continue the driving force suppression control. On the other hand, if accelerator pedal operation amount AP has decreased to or below the termination threshold APE (Yes), ECU 10 proceeds to the process of step S490, terminates the driving force suppression control, and then temporarily ends this routine. Thereafter, ECU 10 repeatedly executes the processes of steps S400 to S490 described above until the ignition switch or start button of vehicle 100 is turned OFF.

[0127] According to the fourth embodiment described above in detail, when a parking row in which the number of consecutive parking stalls PL or parked vehicle contour lines VL is equal to or greater than a first threshold number (e.g., 5) is acquired around the host vehicle 100, and when a temporary parking row in which the number of consecutive stalls PL or parked vehicle contour lines VL is equal to or greater than a second threshold number (e.g., 3) and less than the first threshold number is further acquired, if these parking rows and temporary parking rows are in a predetermined positional relationship, the temporary parking row is determined to be a parking row. This makes it possible to effectively detect, as a parking row, parking stalls PL and parked vehicles 300 arranged in small numbers at the end of an aisle R, for example, in a parking lot P of a large store or the like. Furthermore, when the predicted traveling trajectory TP of the host vehicle 100 intersects with a temporary parking row determined to be a parking row, by enabling the operation of the driving force suppression control, it is possible to reliably improve safety compared to when the control is enabled only for parking rows with a large number of consecutive stalls.

[0128] [others] The above describes the driving assistance device, driving assistance method, and program according to this embodiment, but the present disclosure is not limited to the above embodiment, and various modifications are possible as long as they do not deviate from the purpose of the present invention.

[0129] For example, the above-described embodiments can be combined in any manner. For example, when the second embodiment and the third embodiment are combined, if the determination in step S215 in the control routine of Fig. 19 is negative (No), step S330 in Fig. 25 is executed, and the process proceeds to step S250 in Fig. 19 via the processing of step S338 in Fig. 25.

[0130] When the third embodiment and the fourth embodiment are combined, the control routine of FIG. In this case, the processes of steps S425 to S436 in FIG. 28 may be executed between steps S220 and S250, or between steps S225 and S250.

[0131] Furthermore, in the fourth embodiment, the explanation is given on the premise that a parking row is acquired, but even if a parking row cannot be acquired, it is possible to acquire multiple temporary parking rows and determine these multiple temporary parking rows as parking rows if they are in a predetermined positional relationship. In this case, if the determination in step S415 of FIG. 28 is negative (No), it is determined whether multiple temporary parking rows have been acquired, and if the multiple temporary parking rows are in a predetermined positional relationship, it is sufficient to determine these temporary parking rows as parking rows. [Explanation of symbols]

[0132] 1... driving assistance device, 10... ECU, 11... parking space acquisition unit, 12... parked vehicle acquisition unit, 13A... parking row determination unit, 13B... temporary parking row determination unit, 13C... structure acquisition unit, 14... parking row aisle determination unit, 15... driving trajectory prediction unit, 16... parking lot interior determination unit, 17... erroneous operation determination unit, 18... driving force suppression control unit, 30... vehicle state acquisition device, 40... surrounding recognition device, 41... camera sensor, 42... radar sensor, 20... drive unit, 21... steering device, 21... braking device, 100... host vehicle, 200... lane marking, 300... parked vehicle

Claims

1. a parking row acquisition unit that detects parking spaces and / or parked vehicles around the vehicle based on image data of the surroundings of the vehicle, and acquires a parking row in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and are continuous for a predetermined number of times or more than a first threshold; a parking lot interior determination unit that determines whether the vehicle is present in a parking lot having the parking row; an error operation determination unit that acquires an operation state of an accelerator pedal operated by a driver of the host vehicle to accelerate the host vehicle, and determines whether the driver has erroneously depressed the accelerator pedal based on the operation state; a control unit that executes driving force suppression control to suppress the driving force of the host vehicle when the parking lot interior determination unit determines that the host vehicle is present in the parking lot and the erroneous operation determination unit determines that the driver has performed the erroneous operation; a travel locus prediction unit that predicts a travel locus of the host vehicle based on a travel state of the host vehicle, The parking lot interior determination unit determines that the vehicle is present in the parking lot when the predicted travel path intersects with the parking row. Driving assistance device.

2. The driving assistance device according to claim 1, a parking row passage determination unit that determines an area between the plurality of parking rows as a parking row passage when the parking row acquisition unit acquires the plurality of parking rows that are opposed to each other across the host vehicle and the plurality of parking rows are in a predetermined positional relationship; The parking lot interior determination unit determines that the vehicle is located in the parking lot when the vehicle is located in the area. Driving assistance device.

3. The driving assistance device according to claim 1, a structure acquisition unit that acquires structures present around the host vehicle based on target information around the host vehicle; a temporary parking line acquisition unit that detects parking spaces and / or parked vehicles around the vehicle based on the image data, and acquires a temporary parking line in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being less than the first threshold number; When the structure acquisition unit acquires a structure of a predetermined length or more that exists on the opposite side of the temporary parking row from the vehicle, and when the structure and the temporary parking row are in a predetermined positional relationship, the parking row acquisition unit acquires the temporary parking row as a parking row. Driving assistance device.

4. The driving assistance device according to claim 2, a structure acquisition unit that acquires structures present around the host vehicle based on target information around the host vehicle; a temporary parking line acquisition unit that detects parking spaces and / or parked vehicles around the vehicle based on the image data, and acquires a temporary parking line in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being less than the first threshold number; When the structure acquisition unit acquires a structure of a predetermined length or more that exists on the opposite side of the temporary parking row from the vehicle, and when the structure and the temporary parking row are in a predetermined positional relationship, the parking row acquisition unit acquires the temporary parking row as a parking row. Driving assistance device.

5. The driving assistance device according to claim 1, A temporary parking line acquisition unit is provided which detects parking spaces and / or parked vehicles around the vehicle based on the image data, and acquires temporary parking lines in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being smaller than the first threshold number, When the parking row acquisition unit acquires at least one parking row, if the parking row and the temporary parking row are in a predetermined positional relationship, the temporary parking row is acquired as a parking row. Driving assistance device.

6. The driving assistance device according to claim 2, A temporary parking line acquisition unit is provided which detects parking spaces and / or parked vehicles around the vehicle based on the image data, and acquires temporary parking lines in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and are continuous for a predetermined second threshold number or more, the second threshold number being smaller than the first threshold number, When the parking row acquisition unit acquires at least one parking row, if the parking row and the temporary parking row are in a predetermined positional relationship, the temporary parking row is acquired as a parking row. Driving assistance device.

7. Detecting parking spaces and / or parked vehicles around the vehicle based on image data of the surroundings of the vehicle, and acquiring a parking row in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and continue for a predetermined number of times or more than a predetermined first threshold value; determining whether the vehicle is present in a parking lot having the parking row; acquiring an operation state of an accelerator pedal operated by a driver of the host vehicle to accelerate the host vehicle, and determining whether the driver has erroneously depressed the accelerator pedal based on the operation state; When it is determined that the host vehicle is present in the parking lot and that the erroneous operation has been performed by the driver, a driving force suppression control is executed to suppress the driving force of the host vehicle; predicting a travel trajectory of the host vehicle based on a travel state of the host vehicle; When the predicted travel path intersects with the parking row, the computer executes a process of determining that the vehicle is present in the parking lot. Driving assistance methods.

8. The computer of the driving assistance device Detecting parking spaces and / or parked vehicles around the vehicle based on image data of the surroundings of the vehicle, and acquiring a parking row in which the detected parking spaces and / or parked vehicles are adjacent to each other in a predetermined direction and continue for a predetermined number of times or more than a predetermined first threshold value; determining whether the vehicle is present in a parking lot having the parking row; acquiring an operation state of an accelerator pedal operated by a driver of the host vehicle to accelerate the host vehicle, and determining whether the driver has erroneously depressed the accelerator pedal based on the operation state; When it is determined that the host vehicle is present in the parking lot and that the erroneous operation has been performed by the driver, a driving force suppression control is executed to suppress the driving force of the host vehicle; predicting a travel trajectory of the host vehicle based on a travel state of the host vehicle; When the predicted travel path intersects with the parking row, a process is performed to determine that the vehicle is present in the parking lot. program.

Images