JP2017019435A - Parking assistance device and parking assistance method - Google Patents

Abstract

A parking support device and a parking support method capable of appropriately guiding a host vehicle to an initial parking position are provided.
An initial parking position is set in the vicinity of a target parking frame, an obstacle existing in the vicinity of the host vehicle is detected, a feature point of the detected obstacle is extracted, and matching that matches the extracted feature point is performed. A straight line is detected, and the target traveling direction, which is the direction to guide the vehicle from the current position to the initial parking position, is set according to the detected matching straight line, and the initial parking is started from the current position along the set target traveling direction. Guide your vehicle to the position.
[Selection] Figure 5

Description

  The present invention relates to a parking assistance device and a parking assistance method for providing parking assistance for a driver.

As a technique for providing parking assistance for a driver, for example, there is a technique described in Patent Document 1.
In the technique described in Patent Document 1, obstacles and parking frames around the host vehicle are detected. Then, the host vehicle is guided to a parking space adjacent to the obstacle or an initial parking position where parking in the parking frame is possible based on the detected obstacle or the distance along the vehicle width direction between the parking frame and the host vehicle. To do.

JP 2008-195357 A

In the technique described in Patent Document 1, when the traveling distance at the time of guidance becomes long in a state where the traveling direction of the vehicle is inclined with respect to the traveling direction to the target parking position in plan view, the host vehicle is moved to the initial parking position. The direction of guidance deviates from the traveling direction of the vehicle. For this reason, there is a problem in that the direction in which the host vehicle is guided from the current position to the initial parking position is not clearly set, and the host vehicle may not be properly guided to the initial parking position.
The present invention has been made paying attention to the above problems, and provides a parking support apparatus and a parking support method capable of appropriately guiding the host vehicle from the current position to the initial parking position. For the purpose.

  In order to solve the above problems, according to one aspect of the present invention, a parking initial position is set in the vicinity of a target parking frame, and feature points of detected obstacles existing around the host vehicle are extracted. Then, a target traveling direction that is a direction for guiding the host vehicle from the current position to the initial parking position is set according to a straight line that matches the extracted feature point, and the current position is set along the set target traveling direction. To guide the vehicle to the initial parking position.

According to one aspect of the present invention, feature points of obstacles such as parked vehicles existing around the host vehicle are extracted, a straight line that matches the extraction result is detected, the traveling direction of the host vehicle is identified, It becomes possible to guide the host vehicle from the position to the initial parking position.
This makes it possible to set the traveling direction for guiding the vehicle from the current position to the initial parking position according to the feature points of obstacles existing around the vehicle. It is possible to appropriately guide to the position.

It is a block diagram showing the structure of the parking assistance apparatus of 1st embodiment of this invention. It is a figure showing the structure of a vehicle provided with the parking assistance apparatus of 1st embodiment of this invention. It is a top view showing the process which the parking assistance apparatus of 1st embodiment of this invention performs. It is a top view showing the process which the parking assistance apparatus of 1st embodiment of this invention performs. It is a flowchart showing the operation | movement performed using the parking assistance apparatus of 1st embodiment of this invention. It is a top view showing the process which the parking assistance apparatus of 2nd embodiment of this invention performs. It is a top view showing the process which the parking assistance apparatus of 2nd embodiment of this invention performs. It is a block diagram showing the structure of the parking assistance apparatus of 3rd embodiment of this invention. It is a flowchart showing the process performed using the parking assistance apparatus of 3rd embodiment of this invention.

  In the following detailed description, specific details are set forth in order to provide a thorough understanding of the embodiments of the invention. However, it will be apparent that one or more embodiments may be practiced without such specific details. In other instances, well-known structures and devices are schematically shown in order to simplify the drawing.

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.
(Configuration of parking assistance device)
The structure of the parking assistance apparatus 1 is demonstrated using FIGS. 1-4.
As shown in FIG. 1, the parking assistance device 1 includes a front distance sensor 2F, a right side distance sensor 2R, a left side distance sensor 2L, a movement distance sensor 4, a steering angle sensor 6, and a parking assistance ECU 10. Prepare. In addition to this, the parking assist device 1 includes an HMI device 12, a steering mechanism 14, a driving force controller 16, and a braking force controller 18.

The front distance sensor 2F is formed using, for example, a laser scanner. Further, the front distance sensor 2F is mounted on a vehicle including the parking assist device 1 (in the following description, it may be described as “own vehicle”). Further, as shown in FIG. 2, the front distance sensor 2F is mounted at a position in front of the host vehicle MV in the vehicle front-rear direction (for example, the center of the front bumper).
Further, the front distance sensor 2F irradiates a laser toward a region set in front of the host vehicle MV in the vehicle front-rear direction (which may be referred to as a “front detection region” in the following description). When the front distance sensor 2F receives a laser beam reflected by an obstacle such as another vehicle, the information signal including the transmission time and the reception time of the laser (in the following description, “forward transmission / reception time signal” may be described). Is output to the parking assist ECU 10.
Moreover, in 1st embodiment, the case where the magnitude | size of a front detection area | region is set to the area | region within 10 [m] from the own vehicle MV is demonstrated as an example.

The right side distance sensor 2R is formed by using, for example, a laser scanner in the same manner as the front distance sensor 2F. Further, as shown in FIG. 2, the right side distance sensor 2R is mounted at a position (for example, near the right rear door) on the right side in the vehicle width direction of the host vehicle MV (the right side when the host vehicle MV is moving forward). .
Further, the right side distance sensor 2R irradiates a laser toward a region set to the right side in the vehicle width direction of the host vehicle MV (in the following description, it may be described as a “right side detection region”). When the right distance sensor 2R receives the laser reflected by the obstacle, the information signal including the transmission time and the reception time of the laser (in the following description, it may be described as “right-side transmission / reception time signal”). Is output to the parking assist ECU 10.

In the first embodiment, a case where the size of the right side detection area is set to an area within 10 [m] from the host vehicle MV will be described as an example.
The left side distance sensor 2L is formed by using, for example, a laser scanner, similarly to the front distance sensor 2F. Further, as shown in FIG. 2, the left-side distance sensor 2L is mounted at a position (for example, near the left rear door) on the left side in the vehicle width direction of the host vehicle MV (left side when the host vehicle MV moves forward). .
Further, the left side distance sensor 2L irradiates a laser toward a region set to the left side in the vehicle width direction of the host vehicle MV (in the following description, it may be described as a “left side detection region”). When the left-side distance sensor 2L receives the laser reflected by the obstacle or the like, the left-side distance sensor 2L may be described as an information signal including the laser transmission time and reception time (in the following description, “left-side transmission / reception time signal”). Is output to the parking assist ECU 10.

In the first embodiment, as an example, a case where the size of the left side detection area is set to an area within 10 [m] from the host vehicle MV will be described.
The configuration of the front distance sensor 2F, the right side distance sensor 2R, and the left side distance sensor 2L is not limited to the configuration formed using a laser scanner, and for example, formed using a radar or a stereo camera. Also good.
That is, the configuration of the front distance sensor 2F may be any configuration that can detect the distance between the obstacle present in the front detection area and the host vehicle MV. Similarly, the configuration of the right side distance sensor 2R may be a configuration that can detect the distance between the obstacle present in the right side detection region and the host vehicle MV. Further, the configuration of the left side distance sensor 2L may be a configuration capable of detecting the distance between the obstacle present in the left side detection region and the host vehicle MV.

The movement distance sensor 4 is formed using, for example, a wheel speed sensor 8 installed on the wheel W of the host vehicle MV, as shown in FIG. The configuration of the wheel W and the wheel speed sensor 8 will be described later.
Further, the movement distance sensor 4 calculates the movement amount of the host vehicle MV, and sends an information signal including the calculated movement amount (may be referred to as “movement amount signal” in the following description) to the parking assistance ECU 10. Output.
In addition, the structure of the movement distance sensor 4 is not limited to the structure formed using the wheel speed sensor 8, For example, the own vehicle MV is integrated by integrating the value of the vehicle speed sensor which detects the speed of the own vehicle MV. The amount of movement may be calculated.

For example, as shown in FIG. 2, the steering angle sensor 6 is provided in a steering column (not shown) that rotatably supports the steering wheel SW.
The steering angle sensor 6 detects a current steering angle, which is a current rotation angle (steering operation amount) of the steering wheel SW. Then, an information signal including the detected current steering angle (in the following description, may be described as “current steering angle signal”) is output to the parking assist ECU 10. In addition, you may detect the information signal containing the turning angle of a steered wheel (for example, front wheel WFR, WFL) as information showing a steering angle.

Information signals are output from the front distance sensor 2F, the right side distance sensor 2R, the left side distance sensor 2L, the movement distance sensor 4 and the steering angle sensor 6 to the parking assist ECU 10 via the communication bus CB. As the communication bus CB, for example, a CAN (Controller Area Network) is used.
A parking assistance ECU (Electronic Control Unit) 10 is a controller that comprehensively controls the parking assistance device 1 and is constituted by a microcomputer. Note that the microcomputer includes, for example, a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like.

Further, the parking assist ECU 10 is configured by a program that operates using a microcomputer and a memory, for example.
In the first embodiment, a case where the parking assist ECU 10 is configured by hardware will be described. As hardware which comprises parking assistance ECU10, ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) etc. are used, for example.
However, in FIG. 1, the configuration of the parking assist ECU 10 is shown separately for each functional block for convenience. Accordingly, in the following description, the configuration of the parking assist ECU 10 will be described for each functional block for convenience.
The specific configuration of the parking assist ECU 10 will be described later.

The HMI (Human Machine Interface) device 12 is a device that presents parking assistance information including a guidance route to the driver of the host vehicle MV. The HMI device 12 includes a display device such as a display and various meters, and an audio output device such as a speaker.
Further, the HMI device 12 switches display contents on the display, various meters, etc., and audio contents output from the speakers, according to the information presentation command included in the information signal received from the parking assist ECU 10. The information presentation command will be described later.
The steering mechanism 14 is a mechanism that steers the steering wheel SW by a steering motor (not shown), for example. Further, the steering mechanism 14 drives the steering motor in accordance with a steering command included in the information signal received from the parking assist ECU 10, and steers the steering wheel SW. The steering command will be described later.

The driving force controller 16 controls the driving force of the host vehicle MV and is constituted by a microcomputer.
Further, the driving force controller 16 controls the driving force generated by the driving source (for example, engine) of the host vehicle MV in accordance with the forward command included in the information signal received from the parking assist ECU 10. The forward command will be described later.
The braking force controller 18 controls the braking force of the host vehicle MV and is constituted by a microcomputer.
In addition, the braking force controller 18 controls the braking force generated by a braking force generation source (for example, a hydraulic pressure control device) included in the host vehicle MV in response to a stop command included in the information signal received from the parking assist ECU 10. To do. The stop command will be described later.

(Configuration of own vehicle MV)
With reference to FIGS. 1 to 2, the configuration of the host vehicle MV including the parking assistance device 1 will be described.
As shown in FIG. 2, the host vehicle MV including the parking assist device 1 includes a front distance sensor 2F, a right side distance sensor 2R, a left side distance sensor 2L, a wheel speed sensor 8, and a steering angle sensor 6. The parking assistance ECU 10 is provided. In addition, the host vehicle MV includes an HMI device 12, a steering mechanism 14, a driving force controller 16, a braking force controller 18, and wheels W (right front wheel WFR, left front wheel WFL, right rear wheel WRR, left rear wheel). WRL).
Since the configurations of the front distance sensor 2F, the right side distance sensor 2R, and the left side distance sensor 2L have been described above, description thereof will be omitted.

The wheel speed sensor 8 is provided corresponding to each wheel W.
The wheel speed sensor 8 generates a preset number of wheel speed pulses for one rotation of the corresponding wheel W. Then, the wheel speed sensor 8 outputs an information signal including the generated wheel speed pulse to the movement distance sensor 4.
In FIG. 2, a wheel speed sensor 8 that generates a wheel speed pulse for one rotation of the right front wheel WFR is indicated as a wheel speed sensor 8FR, and a wheel speed sensor 8 that generates a wheel speed pulse for one rotation of the left front wheel WFL. Is represented as a wheel speed sensor 8FL. Similarly, in FIG. 2, a wheel speed sensor 8 that generates a wheel speed pulse for one rotation of the right rear wheel WRR is indicated as a wheel speed sensor 8RR, and a wheel that generates a wheel speed pulse for one rotation of the left rear wheel WRL. The speed sensor 8 is represented as a wheel speed sensor 8RL. In the following description, each wheel W and each wheel speed sensor 8 may be expressed as described above.

Since the configuration of the steering angle sensor 6 has been described above, the description thereof is omitted.
A specific configuration of the parking assist ECU 10 will be described later.
Since the configurations of the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 have been described above, description thereof will be omitted.
The driving force controlled by the driving force controller 16 is applied to each wheel W from a driving source (not shown).
The driving force controlled by the braking force controller 18 is applied to each wheel W through a wheel cylinder (not shown) provided corresponding to each wheel W.

(Configuration of parking assist ECU 10)
With reference to FIG. 1, the structure of parking assistance ECU10 is demonstrated.
As shown in FIG. 1, the parking assistance ECU 10 includes a surrounding environment detection unit 20, a parking initial position setting unit 30, a feature point extraction unit 40, a matching straight line detection unit 50, a target travel direction setting unit 60, A forward path calculation unit 70 and an information output control unit 80 are provided.
The surrounding environment detection unit 20 receives input of information signals from the front distance sensor 2F, the right side distance sensor 2R, the left side distance sensor 2L, and the movement distance sensor 4. And the surrounding environment detection part 20 detects the obstruction which exists around the own vehicle MV according to the received information signal.
Specifically, the laser transmission time and reception time included in the forward transmission / reception time signal and the movement amount of the host vehicle MV included in the movement amount signal are referred to. Then, as shown in FIG. 3, an obstacle such as a parked vehicle PV (another vehicle) existing in the front detection area SEF is detected.

In FIG. 3, as an example, the own vehicle MV enters a parking lot with a plurality of parking frames, and the parking frame exists on the left side of the own vehicle MV (the left side when the own vehicle MV moves forward). A target parking initial position SP is set in the vicinity of. And the case where the own vehicle MV moves in the parking lot toward the set parking initial position SP is shown.
In addition, the surrounding environment detection unit 20 refers to the laser transmission time and reception time included in the right side transmission / reception time signal and the movement amount of the host vehicle MV included in the movement amount signal. Then, as shown in FIG. 3, an obstacle present in the right side detection area SER is detected.

In addition, the surrounding environment detection unit 20 refers to the laser transmission time and reception time included in the left side transmission / reception time signal and the movement amount of the host vehicle MV included in the movement amount signal. Then, as shown in FIG. 3, an obstacle present in the left side detection area SEL is detected.
In FIG. 3, as an example, the front distance sensor 2F, the right side distance sensor 2R, and the left side distance sensor 2L are used as obstacles, such as the front bumper (front bumper) end of the parked vehicle PV or the parked vehicle PV. This represents the case where the side is detected.
The surrounding environment detection unit 20 that detects an obstacle existing around the host vehicle MV uses an information signal including the detected obstacle (in the following description, may be described as “obstacle signal”) as a feature point. Output to the extraction unit 40.

The initial parking position setting unit 30 receives information signals from the front distance sensor 2F, the right side distance sensor 2R, the left side distance sensor 2L, and the movement distance sensor 4. And the parking initial position setting part 30 sets parking initial position SP as represented in FIG. 3 according to the received information signal. The parking initial position SP is a position that is set in the vicinity of the target parking frame (target parking frame TPE), and is, for example, a position at which the host vehicle MV starts to move to the target parking frame.
Specifically, the front transmission / reception time signal, right side transmission / reception time signal, left side transmission / reception time signal, and movement amount signal are used to calculate the width and depth of the target parking frame TPE. Then, the position at which the host vehicle MV starts to move backward toward the target parking frame TPE is set in association with the target parking frame TPE, and the set position is set as the parking initial position SP.

In addition, the parking initial position SP should just be the vicinity of the target parking frame TPE, and is not necessarily limited.
That is, in the first embodiment, as an example, a case will be described in which the parking initial position SP is a position where the host vehicle MV starts to move backward toward the target parking frame TPE.
The position at which the host vehicle MV starts to move backward toward the target parking frame TPE is, for example, the size and shape of the target parking frame TPE, the total length and the entire width of the host vehicle MV, the turning performance of the host vehicle MV, and the target parking frame. It is calculated from the width of the road that continues with TPE. In the present embodiment, as an example, a case will be described in which the parking initial position SP is calculated and set as a position where the host vehicle MV can be retracted into the target parking frame TPE without requiring a turnover operation.

The initial parking position setting unit 30 that has set the initial parking position SP uses an information signal including the set initial parking position SP (in the following description, it may be referred to as “parking initial position signal”) as a forward route calculation unit. Output to 70.
As shown in FIG. 3, the feature point extraction unit 40 sets a plurality of feature points CP of the obstacle included in the obstacle signal in advance according to the obstacle included in the obstacle signal. Extract within. The feature point CP is, for example, a position where the laser irradiated by each distance sensor 2 among obstacles is reflected and the distance sensor 2 receives the reflected laser.

In the first embodiment, as an example, the feature point extraction range ECP1 is a straight line that passes through the center of the host vehicle MV in the plan view and extends in the vehicle front-rear direction, and further from the reference line. The range is within 4 [m] in the vehicle width direction of the vehicle MV. In FIG. 3, the length of the feature point extraction range ECP <b> 1 along the vehicle width direction of the host vehicle MV is represented by a symbol “D <b> 1”. Moreover, in FIG. 3, as an example, when the range of the length D1 from the reference line to the left side when the host vehicle MV moves forward is set as the feature point extraction range ECP1 in the vehicle width direction of the host vehicle MV Represents.
Specifically, only the feature point CP existing in the feature point extraction range ECP is selected from the plurality of feature points CP of the obstacle included in the obstacle signal. Then, the selected feature point CP is extracted as a feature point CP existing in the feature point extraction range ECP.

That is, the feature point extraction unit 40 of the first embodiment has a predetermined distance D1 (feature point extraction range ECP1) along the vehicle width direction of the host vehicle MV among the plurality of feature points CP of the obstacle. ) To extract a plurality of feature points CP.
In FIG. 3, a feature point CP existing in the feature point extraction range ECP1 (in the following description, may be described as “in-range feature point”) is represented by a symbol “CPin”. In addition, a feature point CP that exists outside the feature point extraction range ECP1 (in the following description, may be referred to as “out-of-range feature point”) is represented by a symbol “CPout”.

The feature point extraction unit 40 that has extracted the in-range feature point CPin detects an information signal including the extracted in-range feature point CPin (in the following description, may be referred to as “extracted feature point signal”), and detects a suitable straight line. To the unit 50.
The conforming straight line detection unit 50 selects a straight line that is most suitable for the in-range feature point CPin according to the in-range feature point CPin included in the extracted feature point signal (in the following description, it may be described as “adapted straight line”). To detect.
Specifically, in accordance with the feature point extraction range ECP, out-of-range feature points CPout existing on the side surfaces (both sides in the vehicle width direction) of each parked vehicle PV from the vehicle train group of parked vehicles PV that are obstacles. Remove data. Thereby, as shown in FIG.3 and FIG.4, the position of the bumper (front bumper) edge part of a vehicle front surface is extracted with respect to each parked vehicle PV.

Furthermore, using the extracted position of the front bumper end, a straight line that best matches the in-range feature point CPin existing at each front bumper end is generated. Then, the generated straight line is detected as a matching straight line CL as shown in FIG.
As shown in FIGS. 3 and 4, when the front bumper ends of the parked vehicles PV are aligned in a row when viewed from the own vehicle MV, the conforming straight line CL is detected using the least square method. May be.
The straight line that best matches the in-range feature point CPin is, for example, a straight line that satisfies at least one of the following matching conditions I to III.
Conformity conditions A straight line that best fits the in-range feature point CPin using the in-range feature point CPin extracted while the host vehicle MV moves within a preset distance.
Conformity condition II. A straight line that best fits the in-range feature point CPin using the in-range feature point CPin extracted within a preset time.
Conformity condition III. This is a straight line that matches a preset number of in-range feature points CPin.

That is, in the first embodiment, as an example, the matching line detection unit 50 detects a matching line to a point sequence group formed by the plurality of feature points CP extracted by the feature point extraction unit 40, and sets the matching line CL. A case of detection will be described.
Here, as shown in FIG. 3, the “point sequence group” is formed by a “group” in which “a plurality of feature points CP” existing at the end of the front bumper of the parked vehicle PV are “arrayed”.
The matching line detection unit 50 that has detected the matching line CL outputs an information signal (may be described as “matching line signal” in the following description) including the detected matching line CL to the target traveling direction setting unit 60. To do.

The target travel direction setting unit 60 sets a target travel direction of the host vehicle MV (may be described as “target travel direction” in the following description) in accordance with the suitable straight line CL included in the suitable straight line signal. . That is, the target travel direction setting unit 60 sets a target travel direction, which is a direction for guiding the host vehicle MV from the current position to the parking initial position SP, according to the suitable straight line CL detected by the suitable straight line detection unit 50.
Specifically, as shown in FIG. 4, a straight line that is parallel to the fitting straight line CL included in the fitting straight line signal in plan view and passes through the center in the vehicle width direction of the host vehicle MV is set as the target traveling direction TD. To do. In FIG. 4, for the purpose of explanation, a portion of the target traveling direction TD that is located forward of the host vehicle MV in the vehicle front-rear direction is shown.

The “straight line parallel to the conforming straight line in plan view” includes “a straight line inclined in the range of the conforming straight line and a preset angle in plan view”. In addition, the “range of the straight line and the preset angle in plan view” is set to, for example, a range of +5 [°] to −5 [°] based on the “angle parallel to the fit straight line and plan view” To do.
Therefore, the target travel direction setting unit 60 of the first embodiment, when the inclination angle between the conforming straight line CL and the vehicle front-rear direction of the host vehicle MV is an angle within a preset range with reference to parallelism, The direction of CL is set as the target traveling direction TD.

The target travel direction setting unit 60 that has set the target travel direction TD generates an information signal including the set target travel direction TD (in the following description, it may be described as “target travel direction signal”) as a forward path calculation unit. Output to 70.
The forward route calculation unit 70 refers to the target traveling direction TD included in the target traveling direction signal and the parking initial position SP included in the parking initial position signal. Then, the forward route calculation unit 70 may describe the own vehicle MV from the current position to the parking initial position SP along the target traveling direction TD (in the following description, “target forward route”). Is calculated).

The forward path calculation unit 70 that has calculated the target forward path outputs an information signal including the calculated target forward path (may be described as a “target forward path signal” in the following description) to the information output control unit 80. To do.
The information output control unit 80 generates an information signal to be output to the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 according to the target forward path included in the target forward path signal.
The information signal to be output to the HMI device 12 is generated including an information presentation command that proposes a target forward path to the driver of the host vehicle MV. The target advance route is presented, for example, by displaying an image of the target advance route on the display device or by outputting a sound for promoting traveling along the target advance route from the audio output device.

The information signal output to the steering mechanism 14 is generated including a steering command for guiding the traveling route of the host vehicle MV to the target forward route by automatically steering the steering wheel SW by a steering motor provided in the steering mechanism 14.
The information signal output to the driving force controller 16 is generated including a forward command to advance the host vehicle MV from the current position to the parking initial position SP by controlling the driving force generated by the driving source of the host vehicle MV. To do.
The information signal output to the braking force controller 18 controls the braking force generated by the braking force generation source included in the own vehicle MV, so that the own vehicle MV moving forward toward the parking initial position SP is set at the parking initial position. Generated including a stop command to stop.
Therefore, the forward path calculation unit 70, the information output control unit 80, the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 are set along the target travel direction TD set by the target travel direction setting unit 60. The host vehicle MV is guided to the parking initial position SP.

(Operation)
Next, with reference to FIGS. 1 to 4, an operation performed using the parking assist device 1 of the first embodiment will be described using FIG. 5.
In the operation performed using the parking assist device 1, the host vehicle MV approaches the parking initial position SP in the scene where the host vehicle MV moves to the target parking frame TPE in a parking lot where a plurality of parking frames are provided. The guide route is calculated from the obstacles around. And while showing the guidance path | route from the present position to the parking initial position SP with respect to the driver | operator of the own vehicle MV, the own vehicle MV is made to drive | work automatically from the present position to the parking initial position SP.

As shown in FIG. 5, when the operation performed using the parking assist device 1 is started (START), first, the process of step S100 is performed.
In step S100, the surrounding environment detection unit 20 detects an obstacle existing around the host vehicle MV ("obstacle detection" shown in the figure). In step S100, when an obstacle existing around the host vehicle MV is detected, the operation performed using the parking assist device 1 proceeds to step S102.
In step S102, the parking initial position setting unit 30 sets the parking initial position SP ("parking initial position setting" shown in the figure). When the parking initial position SP is set in step S102, the operation performed using the parking assist device 1 proceeds to step S104.

In step S104, the feature point extraction unit 40 extracts a plurality of feature points CP of the obstacle detected in step S100 in the feature point extraction range ECP (“feature points are extracted in the feature point extraction range” shown in the figure). )). When a plurality of feature points CP are extracted in the feature point extraction range ECP in step S104, the operation performed using the parking assistance device 1 proceeds to step S106.
In step S106, the matching line detection unit 50 detects the matching line CL (“matching line detection” shown in the drawing) in accordance with the in-range feature point CPin extracted in step S104. When the matching straight line CL is detected in step S106, the operation performed using the parking assist device 1 proceeds to step S108.

In step S108, the target travel direction setting unit 60 sets the target travel direction TD ("target travel direction setting" shown in the drawing) according to the matching straight line CL detected in step S106. When the target traveling direction TD is set in step S108, the operation performed using the parking assistance device 1 proceeds to step S110.
In step S110, the forward path calculation unit 70 calculates the target forward path according to the parking initial position SP set in step S102 and the target travel direction TD set in step S108 ("target forward path calculation shown in the figure"). )). When the target forward route is calculated in step S110, the operation performed using the parking assist device 1 proceeds to step S112.

In step S112, the information output control unit 80 generates an information signal including an information presentation command, a steering command, and a forward command according to the target forward path calculated in step S110. Thereby, in step S112, the own vehicle MV is controlled so that the own vehicle MV approaches the parking initial position SP from the current position ("own vehicle progress control" shown in the drawing).
Specifically, the video of the target forward route is displayed on the display device included in the HMI device 12, and the HMI device 12 outputs a sound prompting traveling along the target forward route from the audio output device, whereby the host vehicle MV The guidance route is presented to the driver.

In addition to this, the steering wheel SW is automatically steered by the steering motor provided in the steering mechanism 14, and the driving force generated by the drive source of the host vehicle MV is controlled to automatically move the host vehicle MV toward the parking initial position SP. Let it run.
In step S112, when the host vehicle MV is controlled so that the host vehicle MV approaches the parking initial position SP from the current position, the operation performed using the parking assist device 1 proceeds to step S114.
In step S114, for example, a process of determining whether or not the position of the host vehicle MV is in the vicinity of the initial parking position SP using data included in information signals received from the front distance sensor 2F and the movement distance sensor 4. ("Own vehicle is in the vicinity of the initial parking position" shown in the figure).

Note that “the vicinity of the initial parking position SP” to be determined in step S114 means that the host vehicle MV traveling toward the initial parking position SP with respect to the center of the initial parking position SP is the initial parking position SP. The range of distance that can be stopped by. Further, “the vicinity of the parking initial position SP” to be determined in step S114 is set according to the speed of the host vehicle MV and the braking performance of the host vehicle MV.
If it is determined in step S114 that the position of the host vehicle MV is in the vicinity of the parking initial position SP (“Yes” shown in the drawing), the operation performed using the parking assist device 1 proceeds to step S112.

On the other hand, if it is determined in step S114 that the position of the host vehicle MV has not reached the vicinity of the parking initial position SP ("No" shown in the figure), the operation performed using the parking assist device 1 proceeds to step S116. Transition.
In step S116, the information output control unit 80 generates an information signal including a stop command according to the target forward path calculated in step S110. Thereby, in step S116, the own vehicle MV is controlled so as to stop the own vehicle MV at the parking initial position SP ("own vehicle stop control" shown in the drawing). Specifically, the steering wheel SW is automatically steered by the steering motor provided in the steering mechanism 14, and the braking force generated by the braking force generation source provided in the host vehicle MV is controlled, so that the host vehicle MV is set at the parking initial position SP. Stop.

In step S116, when the host vehicle MV is controlled so that the host vehicle MV is stopped at the parking initial position SP, the operation performed using the parking assist device 1 is ended (END).
If it is the parking assistance apparatus 1 of 1st embodiment, as mentioned above, it is finally possible to guide the own vehicle MV to the parking initial position SP accurately. Therefore, the parking assist device 1 of the first embodiment can move the host vehicle MV to the vicinity of the target parking position SP while automatically driving the host vehicle MV even in a parking lot where no white line or the like is drawn. It is.
The forward path calculation unit 70, the information output control unit 80, the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 described above correspond to the guiding unit.

Further, the information output control unit 80, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 described above correspond to the automatic travel device provided in the guide unit.
Moreover, as mentioned above, in the parking assistance method implemented by the operation of the parking assistance device 1 of the first embodiment, the parking initial position SP is set, and obstacles existing around the host vehicle MV are detected and detected. The feature point CP of the obstacle is extracted. Then, a matching straight line CL that matches the extracted feature point CP is detected, a target traveling direction TD is set according to the detected matching straight line CL, and a parking initial position is determined from the current position along the set target traveling direction TD. The vehicle MV is guided to SP.
The above-described first embodiment is an example of the present invention, and the present invention is not limited to the above-described first embodiment, and the present invention may be applied to other forms than this embodiment. Various modifications can be made according to the design or the like as long as they do not depart from the technical idea.

(Effects of the first embodiment)
If it is the parking assistance apparatus 1 of 1st embodiment, it will become possible to show the effect described below.
(1) The matching straight line detection unit 50 detects a matching straight line CL that matches the feature point CP extracted by the feature point extraction unit 40, and the target traveling direction setting unit 60 detects the host vehicle from the current position to the parking initial position SP. A target traveling direction TD, which is a direction for guiding MV, is set according to the matching straight line CL. Then, the forward path calculation unit 70, the information output control unit 80, the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 move the host vehicle MV from the current position along the target traveling direction TD. Guide to the initial parking position SP.

For this reason, feature points CP of obstacles such as a parked vehicle PV existing around the host vehicle MV are extracted, a matching straight line CL that matches the extraction result is detected, and the traveling direction of the host vehicle MV is specified. It is possible to guide the host vehicle MV from the position to the parking initial position SP.
As a result, the traveling direction for guiding the host vehicle MV from the current position to the parking initial position SP can be set according to the feature point CP of the obstacle present around the host vehicle MV. Accordingly, it is possible to appropriately guide the host vehicle MV from the current position to the target parking position SP.

(2) The matching straight line detection unit 50 detects the straight line that best matches the feature point CP extracted by the feature point extraction unit 40 as the matching straight line CL.
For this reason, the fitting straight line CL that best matches the feature point CP extracted by the feature point extraction unit 40 is detected from the extraction result of the feature point CP set in the obstacle such as the parked vehicle PV existing around the host vehicle MV. It becomes possible.
As a result, the traveling direction of the host vehicle MV that best matches the feature point CP extracted by the feature point extraction unit 40 can be specified, and the host vehicle MV can be guided from the current position to the parking initial position SP. .

(3) The feature point extraction unit 40 extracts a plurality of feature points CP existing in the feature point extraction range ECP1 of the distance D1 set in advance along the vehicle width direction of the host vehicle MV. Furthermore, the matching line detection unit 50 detects the matching line CL that matches the point sequence group formed by the plurality of feature points CP extracted by the feature point extraction unit 40. In addition, the target travel direction setting unit 60 detects when the tilt angle between the target straight line CL and the vehicle front-rear direction of the host vehicle MV is an angle within a preset range with reference to parallel. The direction of the adapted straight line CL is set as the target traveling direction TD.

Therefore, the inclination angle between the matching straight line CL detected according to the point sequence group formed by the plurality of feature points CP that are close in the vehicle width direction of the host vehicle MV and the vehicle front-rear direction of the host vehicle MV is A straight line that is substantially parallel can be set as the target traveling direction TD.
As a result, it is possible to set the traveling direction for guiding the host vehicle MV from the current position to the parking initial position SP according to a plurality of feature points CP that are close in the vehicle width direction of the host vehicle MV. It becomes. Accordingly, it is possible to appropriately guide the host vehicle MV from the current position to the target parking position SP.

Further, in the above-described conventional technology, linear approximation or curve approximation is performed for all point sequences or point sequences longer than a predetermined length based on the point sequence data representing the distance between the obstacle and the host vehicle MV. I do. For this reason, in the prior art, when a point group outside the detection range is detected, the position of the line approximated to all point sequences or point sequences of a predetermined length or more is the actual obstacle position. There is a possibility that the difference between the direction in which the host vehicle MV is actually guided and the target traveling direction TD increases.
On the other hand, in the first embodiment, the feature point extraction unit 40 extracts a plurality of feature points CP existing in the feature point extraction range ECP1, and the matching line detection unit 50 extracts a plurality of features extracted by the feature point extraction unit 40. The matching straight line CL that matches the point sequence group formed by the feature points CP is detected. For this reason, it is possible to reduce the error between the direction in which the host vehicle MV is actually guided and the target traveling direction TD.

(4) The automatic travel device formed by using the information output control unit 80, the steering mechanism 14, the driving force controller 16, and the braking force controller 18 automatically causes the host vehicle MV to travel from the current position to the parking initial position SP.
As a result, it is not necessary for the driver to move the host vehicle MV to the parking initial position SP as in the prior art. As a result, it is not necessary for the driver to be aware of the position at which driving assistance control such as automatic steering is started, so that it is possible to greatly reduce the driver's effort.

(5) The parking initial position SP is set to a position at which the host vehicle MV starts moving backward toward the target parking frame (target parking frame TPE).
As a result, the host vehicle MV guided to the parking initial position SP can be moved to the target parking frame TPE only by traveling backward. Thereby, it is possible to reduce the load that the driver receives during driving. Moreover, when starting from the target parking frame TPE, it is possible to reduce the load that the driver receives during driving.

(6) In the parking assistance method implemented by the operation of the parking assistance device 1 of the first embodiment, the parking initial position SP is set, obstacles existing around the host vehicle MV are detected, and the features of the detected obstacles are detected. A point CP is extracted. Then, a matching straight line CL that matches the extracted feature point CP is detected, a target traveling direction TD is set according to the detected matching straight line CL, and a parking initial position is determined from the current position along the set target traveling direction TD. The vehicle MV is guided to SP.
For this reason, feature points CP of obstacles such as a parked vehicle PV existing around the host vehicle MV are extracted, a matching straight line CL that matches the extraction result is detected, and the traveling direction of the host vehicle MV is specified. It is possible to guide the host vehicle MV from the position to the parking initial position SP.
As a result, the traveling direction for guiding the host vehicle MV from the current position to the parking initial position SP can be set according to the feature point CP of the obstacle present around the host vehicle MV. Accordingly, it is possible to appropriately guide the host vehicle MV from the current position to the target parking position SP.

(Modification of the first embodiment)
(1) In the first embodiment, the configuration of the parking assist device 1 automatically controls the steering wheel SW, controls the driving force and the braking force, and automatically moves the host vehicle MV from the current position to the parking initial position SP. The configuration is such that automatic movement is performed. However, the structure of the parking assistance apparatus 1 is not limited to this.
That is, the configuration of the parking assistance device 1 may be a configuration in which, for example, the HMI device 12 only presents the guidance route to the driver of the host vehicle MV. Further, the configuration of the parking assist device 1 may be configured to perform only automatic steering of the steering wheel SW, for example. In addition, the configuration of the parking assist device 1 may be configured to perform only driving operation assistance such as steering assist of the steering wheel SW, assistance of braking force and driving force, and the like.
Therefore, the configuration of the parking assist device 1 may be configured to include at least one of the HMI device 12, the steering mechanism 14, the driving force controller 16, and the braking force controller 18.

(2) In the first embodiment, the matching straight line detection unit 50 detects the straight line that best matches the feature point CP extracted by the feature point extraction unit 40 as the matching line CL. However, the present invention is not limited to this. In other words, the matching straight line detection unit 50 may detect the second or third matching straight line with respect to the feature point CP extracted by the feature point extraction unit 40 as the matching straight line CL.
(3) In the first embodiment, the configurations of the front distance sensor 2F, the right side distance sensor 2R, and the left side distance sensor 2L are formed using a laser scanner, but are not limited thereto. That is, the configuration of the front distance sensor 2F, the right side distance sensor 2R, and the left side distance sensor 2L may be formed using, for example, an ultrasonic sonar. In this case, when passing near the obstacle, the detection result of the obstacle existing around the host vehicle MV may be recorded in the memory using the movement amount calculated by the movement distance sensor 4.

(4) In the first embodiment, as shown in FIGS. 3 and 4, the case where the target parking position (target parking frame TPE) is set on the left side when the host vehicle MV moves forward and parking is performed has been described. However, the present invention is not limited to this. That is, the present invention may be applied when parking is performed by setting a target parking position (target parking frame TPE) on the right side when the host vehicle MV moves forward.
Even when the target parking position is set on the right side when the host vehicle MV moves forward, obstacles existing around the host vehicle MV can be avoided by performing the same processing as in the first embodiment described above. However, it is possible to guide the host vehicle MV to the parking initial position SP.
In addition, the configuration of the parking assist device 1 is detected in advance from the vehicle group of the parked vehicles PV existing on both sides in the vehicle width direction of the host vehicle MV by detecting parked vehicles PV positioned on both sides of the host vehicle MV in the vehicle width direction. The target traveling direction TD may be set.

(Second embodiment)
Hereinafter, a second embodiment of the present invention will be described with reference to the drawings.
(Configuration of parking assistance device)
The structure of the parking assistance apparatus 1 is demonstrated using FIG.6 and FIG.7, referring FIGS. 1-5. In addition, the description about the structure similar to 1st embodiment mentioned above may be abbreviate | omitted.
The configuration of the parking assistance device 1 of the second embodiment is the same as that of the first embodiment described above, except for the configuration of the feature point extraction unit 40, the adapted straight line detection unit 50, and the target travel direction setting unit 60. Therefore, in the following description, the description of the components other than the feature point extraction unit 40, the conforming straight line detection unit 50, and the target travel direction setting unit 60 may be omitted.
As shown in FIG. 6, the feature point extraction unit 40 sets a plurality of feature points CP of the obstacle included in the obstacle signal within the preset feature point extraction range ECP2 according to the obstacle included in the obstacle signal. Extract with

In the second embodiment, as an example, the feature point extraction range ECP2 is obtained by using a straight line passing through the center of the rear wheels WRL and WRR of the host vehicle MV as a reference line in plan view, and further using the reference point as a center. A range having a width within 0.5 [m] in the vehicle longitudinal direction of the vehicle MV. In FIG. 6, the length of the feature point extraction range ECP2 along the vehicle front-rear direction of the host vehicle MV is represented by reference sign “D2”.
Specifically, only the feature point CP existing in the feature point extraction range ECP2 is selected from the plurality of feature points CP of the obstacle included in the obstacle signal. Then, the selected feature point CP is extracted as a feature point CP existing in the feature point extraction range ECP.

That is, the feature point extraction unit 40 of the second embodiment includes a range of the distance D2 in which the distance along the vehicle front-rear direction of the host vehicle MV among the plurality of feature points CP of the obstacle (feature point extraction range ECP2 ) To extract a plurality of feature points CP.
In FIG. 6, a feature point CP existing in the feature point extraction range ECP <b> 2 (may be described as “in-range feature point” in the following description) is represented by a symbol “CPin”. In addition, a feature point CP that exists outside the feature point extraction range ECP2 (in the following description, may be described as “out-of-range feature point”) is represented by a symbol “CPout”.

The feature point extraction unit 40 that has extracted the in-range feature point CPin outputs the extracted feature point signal to the matching line detection unit 50.
The matching straight line detection unit 50 detects the matching straight line CL according to the in-range feature point CPin included in the extracted feature point signal.
Specifically, according to the feature point extraction range ECP2, one parked vehicle PV existing in the feature point extraction range ECP2 is detected from the vehicle train group of each parked vehicle PV that is an obstacle. Then, for the one parked vehicle PV existing in the feature point extraction range ECP2, the data of the out-of-range feature point CPout existing at the bumper (front bumper) end of the vehicle front is removed. As a result, as shown in FIGS. 6 and 7, the position of the right side surface of the vehicle is extracted with respect to one parked vehicle PV existing in the feature point extraction range ECP2.

Furthermore, a straight line that best matches the in-range feature point CPin existing on the vehicle right side surface is generated using the extracted position of the vehicle right side surface. Then, the generated straight line is detected as a conforming straight line CL as shown in FIG.
As shown in FIGS. 6 and 7, the left side distance sensor 2L, that is, one distance sensor 2 can detect the side surface of one parked vehicle PV existing in the feature point extraction range ECP2. Alternatively, the fitting straight line CL may be detected using a least square method.
That is, in the second embodiment, as an example, the matching line detection unit 50 detects a matching line to a point sequence group formed by the plurality of feature points CP extracted by the feature point extraction unit 40, and sets the matching line CL. A case of detection will be described.

Here, as shown in FIG. 6, the “point sequence group” refers to “a plurality of feature points CP existing on the right side of the vehicle with respect to one parked vehicle PV existing in the feature point extraction range ECP2. "Is a" group "of" sequence ".
The matching straight line detection unit 50 that has detected the matching straight line CL outputs a matching straight line signal to the target traveling direction setting unit 60.
The target traveling direction setting unit 60 sets the target traveling direction according to the matching straight line CL included in the matching straight line signal. That is, the target travel direction setting unit 60 sets a target travel direction, which is a direction for guiding the host vehicle MV from the current position to the parking initial position SP, according to the suitable straight line CL detected by the suitable straight line detection unit 50.

Specifically, as shown in FIG. 7, a straight line perpendicular to the matching straight line CL included in the matching straight line signal in plan view is set as the target traveling direction auxiliary line SCL. In FIG. 7, the inclination angle between the matching straight line CL and the target traveling direction auxiliary line SCL is represented by a symbol “θ” in plan view. Further, a straight line that is parallel to the target traveling direction auxiliary line SCL in plan view and passes through the center of the host vehicle MV in the vehicle width direction is set as the target traveling direction TD. In FIG. 7, for the sake of explanation, a portion of the target traveling direction TD that is located in front of the host vehicle MV in the vehicle front-rear direction is shown.
The “straight line that is perpendicular to the conforming straight line in plan view” includes “a straight line that is tilted within a range of angles that are preset with the conforming straight line in plan view”. Further, the “range of the straight line and the preset angle in the plan view” is set to, for example, a range of +5 [°] to −5 [°] based on the “angle perpendicular to the fit straight line and the plan view”. To do.
Therefore, the target travel direction setting unit 60 of the second embodiment is configured so that when the inclination angle between the matching straight line CL and the vehicle front-rear direction of the host vehicle MV is an angle within a preset range with respect to the vertical, the matching straight line The direction of CL is set as the target traveling direction TD.
The target travel direction setting unit 60 that has set the target travel direction TD outputs a target travel direction signal to the forward path calculation unit 70.

(Vehicle configuration)
Since the structure of the own vehicle MV provided with the parking assistance apparatus 1 is the same as that of 1st embodiment mentioned above, the description is abbreviate | omitted.
(Operation)
Since the operation | movement performed using the parking assistance apparatus 1 of 2nd embodiment is the same as that of 1st embodiment mentioned above, the description is abbreviate | omitted.
The above-described second embodiment is an example of the present invention, and the present invention is not limited to the above-described second embodiment, and the present invention may be applied to other forms than this embodiment. Various modifications can be made according to the design or the like as long as they do not depart from the technical idea.

(Effect of the second embodiment)
If it is the parking assistance apparatus 1 of 2nd embodiment, it will become possible to show the effect described below.
(1) The feature point extraction unit 40 extracts a plurality of feature points CP existing in the feature point extraction range ECP2 having a distance D2 set in advance in the vehicle longitudinal direction of the host vehicle MV. Furthermore, the matching line detection unit 50 detects the matching line CL that matches the point sequence group formed by the plurality of feature points CP extracted by the feature point extraction unit 40. In addition to this, the target travel direction setting unit 60 detects when the tilt angle between the target straight line CL and the vehicle width direction of the host vehicle MV is an angle within a preset range with respect to the vertical. The direction of the adapted straight line CL is set as the target traveling direction TD.
For this reason, the fitting straight line CL detected according to the point sequence group formed by the plurality of feature points CP existing in the narrow area along the vehicle longitudinal direction of the host vehicle MV, and the vehicle width direction of the host vehicle MV It is possible to set a straight line in which the inclination angle is substantially vertical as the target traveling direction TD.

As a result, even when the number of parked vehicles is small, it is possible to set the traveling direction for guiding the host vehicle MV from the current position to the parking initial position SP. Accordingly, it is possible to appropriately guide the host vehicle MV from the current position to the target parking position SP.
Further, in the above-described conventional technology, linear approximation or curve approximation is performed for all point sequences or point sequences longer than a predetermined length based on the point sequence data representing the distance between the obstacle and the host vehicle MV. I do. For this reason, in the prior art, when a point group outside the detection range is detected, the position of the line approximated to all point sequences or point sequences of a predetermined length or more is the actual obstacle position. There is a possibility that the difference between the direction in which the host vehicle MV is actually guided and the target traveling direction TD increases.
On the other hand, in the second embodiment, the feature point extraction unit 40 extracts a plurality of feature points CP existing in the feature point extraction range ECP2, and the matching line detection unit 50 extracts a plurality of features extracted by the feature point extraction unit 40. The matching straight line CL that matches the point sequence group formed by the feature points CP is detected. For this reason, it is possible to reduce the error between the direction in which the host vehicle MV is actually guided and the target traveling direction TD.

(Third embodiment)
Hereinafter, a third embodiment of the present invention will be described with reference to the drawings.
(Configuration of parking assistance device)
The configuration of the parking assistance device 1 will be described with reference to FIGS. 1 to 7 and FIGS. 8 and 9. In addition, the description about the structure similar to 1st embodiment mentioned above may be abbreviate | omitted.
The configuration of the parking assistance device 1 of the third embodiment is the same as that of the first embodiment described above except for the configuration of the conforming straight line detection unit 50. Therefore, in the following description, the description of the configuration other than the conforming straight line detection unit 50 may be omitted.
The fitting line detection unit 50 includes a data extraction unit 52, a mathematical model calculation unit 54, and a matching degree calculation unit 56.
The data extraction unit 52 sets a minimum number of feature point data necessary for matching with the mathematical model from the feature point data that is the feature point CP data extracted by the feature point extraction unit 40 for a preset time. Extraction is performed at intervals (for example, 50 [msec]).

The mathematical model calculation unit 54 calculates a mathematical model based on the minimum number of feature point data extracted by the data extraction unit 52.
The degree-of-fit calculation unit 56 calculates the number of feature point data matching the mathematical model calculated by the mathematical model calculation unit 54 from the minimum number of feature point data extracted by the data extraction unit 52.
In addition to this, the degree-of-fit calculation unit 56 sequentially matches all the feature point data extracted by the feature point extraction unit 40 with the formula model calculated by the formula model calculation unit 54. Then, a mathematical model in which the number of feature point data matching the mathematical model calculated by the mathematical model computation unit 54 is equal to or greater than a preset adaptation threshold is detected as the conforming straight line CL. In addition, a suitable threshold value is set according to the performance etc. of parking assistance ECU10.

That is, the conforming straight line detection unit 50 of the third embodiment performs the process shown in FIG. 9 to detect the conforming straight line CL. Note that FIG. 9 shows processing performed by the conforming straight line detection unit 50 and the target travel direction setting unit 60.
As shown in FIG. 9, when the process performed by the matching straight line detection unit 50 and the target travel direction setting unit 60 is started (START), first, the process of step S200 is performed.
In step S200, the data extraction unit 52 randomly extracts n feature point data from the feature point data that is the data of the feature point CP extracted by the feature point extraction unit 40 (“random” shown in the figure). n data is extracted "). Note that “n” is the minimum number necessary for matching with the mathematical model. When n feature point data are extracted at random in step S200, the process performed by the matching line detection unit 50 and the target traveling direction setting unit 60 proceeds to step S202.

In step S202, the mathematical model is calculated by the mathematical model calculation unit 54, and the degree of adaptation between the mathematical model and the feature point data is calculated by the fitness degree calculation unit 56 ("math formula model and fitness is calculated" shown in the figure). )). When the mathematical model and the degree of matching are calculated in step S202, the processing performed by the matching line detection unit 50 and the target travel direction setting unit 60 proceeds to step S204.
In step S204, a score indicating the correctness of the parameters of the mathematical formula model calculated in step S202 by using only the feature point data within a preset range (inlier range) by the fitness degree calculation unit 56 (fitness score) ) Is calculated. That is, in step S204, the fitness score of the parameter (model parameter) of the mathematical model is calculated (“calculate the fitness score of the model parameter” shown in the figure). In step S204, when the fitness score of the model parameter is calculated, the processing performed by the matching line detection unit 50 and the target travel direction setting unit 60 proceeds to step S206.

In step S206, it is determined whether or not the fitness score calculated in step S204 is equal to or greater than a preset fitness threshold (within an allowable range) ("score is within allowable range" shown in the figure). I do.
If it is determined in step S206 that the fitness score is within the allowable range (“Yes” shown in the figure), the processing performed by the fitness line detection unit 50 and the target travel direction setting unit 60 proceeds to step S208.
On the other hand, if it is determined in step S206 that the fitness score is out of the allowable range ("No" shown in the figure), the process performed by the matching line detection unit 50 and the target travel direction setting unit 60 proceeds to step S200. To do.

In step S208, the matching degree calculation unit 56 uses the feature point data extracted in step S200 to determine the inclination angle between the matching straight line CL and the vehicle longitudinal direction of the host vehicle MV or the vehicle width direction of the host vehicle MV. Calculate ("Calculate tilt angle" shown in the figure). In step S208, when the inclination angle between the conforming straight line CL and the vehicle longitudinal direction of the host vehicle MV or the vehicle width direction of the host vehicle MV is calculated, the processing performed by the conforming straight line detection unit 50 and the target traveling direction setting unit 60 is performed. The process proceeds to step S210.
In step S210, a process of determining whether or not the tilt angle calculated in step S208 is within a preset angle threshold ("tilt angle is within angle threshold" shown in the figure) is performed.

In addition, when the inclination angle between the conforming straight line CL and the vehicle front-rear direction of the host vehicle MV is calculated in step S208, in step S210, the angle threshold is set based on “an angle parallel to the conforming straight line CL in plan view”. For example, it is set in the range of +5 [°] to −5 [°]. Further, when the inclination angle between the matching straight line CL and the vehicle width direction of the host vehicle MV is calculated in step S208, in step S210, the angle threshold is set based on “an angle perpendicular to the matching straight line CL in plan view”. For example, it is set in the range of +5 [°] to −5 [°].
If it is determined in step S210 that the inclination angle calculated in step S208 is within the angle threshold (“Yes” shown in the figure), the process performed by the matching line detection unit 50 and the target traveling direction setting unit 60 is step S212. Migrate to

On the other hand, if it is determined in step S210 that the inclination angle calculated in step S208 is not within the angle threshold ("No" shown in the figure), the process performed by the matching line detection unit 50 and the target travel direction setting unit 60 is as follows. The process proceeds to step S200.
In step S212, the target travel direction setting unit 60 updates the target travel direction TD set in the previous process using the inclination angle calculated in step S208 ("update target travel direction" shown in the figure). In step S212, when the target traveling direction TD is updated, the process performed by the matching straight line detecting unit 50 and the target traveling direction setting unit 60 is ended (END).
The above-described processing from step S200 to step S212 is repeated until the evaluation function becomes equal to or less than the threshold value in order to increase the accuracy of the inclination of the straight line for detecting the arrangement of the vehicle train group of the parked vehicle PV. That is, in the third embodiment, the matching straight line CL is detected and the target traveling direction TD is set using a known RANSAC (RANdom Sampl Consensus) method.

(Vehicle configuration)
Since the structure of the own vehicle MV provided with the parking assistance apparatus 1 is the same as that of 1st embodiment mentioned above, the description is abbreviate | omitted.
(Operation)
Since the operation | movement performed using the parking assistance apparatus 1 of 3rd embodiment is the same as that of 1st embodiment mentioned above, the description is abbreviate | omitted.
The above-described third embodiment is an example of the present invention, and the present invention is not limited to the above-described third embodiment, and the present invention is not limited to the above-described third embodiment. Various modifications can be made according to the design or the like as long as they do not depart from the technical idea.

(Effect of the third embodiment)
If it is the parking assistance apparatus 1 of 3rd embodiment, it will become possible to show the effect described below.
(1) The degree-of-fit calculation unit 56 calculates the number of feature point data matching the mathematical model calculated by the mathematical model calculation unit 54 from the minimum number of feature point data extracted by the data extraction unit 52. In addition to this, the degree-of-fit calculation unit 56 sequentially matches all the feature point data extracted by the feature point extraction unit 40 with the formula model calculated by the formula model calculation unit 54. Then, a mathematical model in which the number of feature point data that conforms to the mathematical model calculated by the mathematical model computation unit 54 is greater than or equal to the conformance threshold is detected as the conforming straight line CL.

For this reason, it becomes possible to improve the detection accuracy of the conforming straight line CL as compared with the configuration in which the conforming straight line CL is detected using the first square method.
As a result, the traveling direction for guiding the host vehicle MV from the current position to the parking initial position SP can be set according to the feature point CP set for the obstacle existing around the host vehicle MV. . Accordingly, it is possible to appropriately guide the host vehicle MV from the current position to the target parking position SP.

  DESCRIPTION OF SYMBOLS 1 ... Parking assistance apparatus, 2F ... Forward distance sensor, 2R ... Right side distance sensor, 2L ... Left side distance sensor, 4 ... Movement distance sensor, 6 ... Steering angle sensor, 8 ... Wheel speed sensor, 10 ... Parking assistance ECU, DESCRIPTION OF SYMBOLS 12 ... HMI apparatus, 14 ... Steering mechanism, 16 ... Driving force controller, 18 ... Braking force controller, 20 ... Ambient environment detection part, 30 ... Parking initial position setting part, 40 ... Feature point extraction part, 50 ... Conformity straight line detection part 52 ... Data extraction unit, 54 ... Mathematical model calculation unit, 56 ... Fitness calculation unit, 60 ... Target travel direction setting unit, 70 ... Advance path calculation unit, 80 ... Information output control unit, MV ... Self vehicle, PV ... Parked vehicle, SW ... steering wheel, CB ... communication bus, W ... wheel (left front wheel WFL, right front wheel WFR, left rear wheel WRL, right rear wheel WRR), SP ... parking initial position, TPE ... target parking frame, SEF Front detection area, SER ... right side detection area, SEL ... left side detection area, CP ... feature point, ECP ... feature point extraction range, CPin ... in-range feature point, CPout ... out-of-range feature point, CL ... conforming straight line, TD ... Target traveling direction, SCL ... Target traveling direction auxiliary line

Claims (8)

An initial parking position setting unit for setting an initial parking position in the vicinity of the target parking frame;
A surrounding environment detection unit for detecting obstacles existing around the vehicle;
A feature point extraction unit for extracting feature points of obstacles detected by the surrounding environment detection unit;
A matching straight line detection unit that detects a straight line matching the feature point extracted by the feature point extraction unit;
A target advancing direction setting unit that sets a target advancing direction that is a direction for guiding the host vehicle from the current position to the initial parking position according to the straight line detected by the conforming straight line detecting unit;
A parking assistance device comprising: a guidance unit that guides the host vehicle from the current position to the parking initial position along a target travel direction set by the target travel direction setting unit.   The parking assist device according to claim 1, wherein the matching straight line detection unit detects a straight line that best matches the feature point extracted by the feature point extraction unit. The feature point extraction unit includes a plurality of feature points of the obstacle detected by the surrounding environment detection unit, and a plurality of distances along a vehicle width direction of the host vehicle are present within a preset distance range. Extract feature points,
The conforming line detection unit detects a line conforming to a point sequence group formed by a plurality of feature points extracted by the feature point extraction unit,
The target travel direction setting unit is configured such that when the inclination angle between the straight line detected by the adaptive straight line detection unit and the vehicle front-rear direction of the host vehicle is an angle within a range set in advance with reference to parallel, the adaptive straight line The parking assistance device according to claim 1 or 2, wherein a direction of a straight line detected by the detection unit is set as the target traveling direction. The feature point extraction unit includes a plurality of the feature points of the obstacle detected by the surrounding environment detection unit, and a plurality of distances along a vehicle front-rear direction of the host vehicle exist within a preset distance range. Extract feature points,
The conforming line detection unit detects a line conforming to a point sequence group formed by a plurality of feature points extracted by the feature point extraction unit,
When the inclination angle between the straight line detected by the conforming straight line detection unit and the vehicle width direction of the host vehicle is an angle within a preset range with reference to the vertical, the target traveling direction setting unit is the conforming straight line The parking assistance device according to claim 1 or 2, wherein a direction of a straight line detected by the detection unit is set as the target traveling direction. The conforming line detection unit is configured to extract a minimum number of feature point data necessary for conformity with the mathematical model from feature point data that is feature point data extracted by the feature point extraction unit at a preset time interval. A data extraction unit to extract, a mathematical model calculation unit to calculate the mathematical model based on the minimum number of feature point data extracted by the data extraction unit, and the minimum number extracted by the data extraction unit A degree-of-fit calculation unit that calculates the number of feature point data that matches the mathematical model calculated by the mathematical model calculation unit from the feature point data of
The degree-of-fit calculation unit matches all the feature point data extracted by the feature point extraction unit with the formula model calculated by the formula model calculation unit in order, and matches the formula model calculated by the formula model calculation unit. The parking assistance device according to claim 3 or 4, wherein a mathematical model in which the number of feature point data is greater than or equal to a preset matching threshold value is detected as the matching straight line.   The parking support according to any one of claims 1 to 5, wherein the guide unit includes an automatic travel device that automatically travels the host vehicle from the current position to the initial parking position. apparatus.   The parking support device according to any one of claims 1 to 6, wherein the parking initial position is a position at which the host vehicle starts to move backward toward the target parking frame. Set the parking initial position near the target parking frame,
Detect obstacles around the vehicle,
Extracting feature points of the detected obstacle,
Detecting a straight line that matches the extracted feature points;
A target traveling direction, which is a direction for guiding the host vehicle from the current position to the initial parking position, is set according to the detected straight line,
A parking assistance method, wherein the host vehicle is guided from the current position to the parking initial position along the set target traveling direction.

Images