JP2009290788A - Parking support device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a parking support device capable of more accurately guiding a vehicle into a parking frame line without increasing the size or the like of the system. <P>SOLUTION: The parking support device includes: an imaging means for taking an image of the area behind the relevant vehicle; an on-image target parking position setting means for setting a target parking position in accordance with the parking frame line whose image is taken in the image taken by the imaging means; a target guided position setting means for carrying out predetermined conversion processing on the target parking position set by the on-image target parking position setting means to determine a target guided position; and a guiding means for guiding the vehicle to the target guided position determined by the target guided position setting means. The target guided position setting means determines a target guided position by: correcting the target parking position based on the position and shape of the parking frame line in the image taken by the imaging means; and carrying out predetermined conversion processing on the corrected position. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a parking assist device that guides a vehicle to a target parking position according to a captured image of a camera that captures the rear of the vehicle.

  Conventionally, a control device that performs automatic steering control or the like to guide a vehicle to a target parking position set on an operation screen that displays a captured image of a camera that captures the rear of the vehicle has been put into practical use (for example, non-patent) Reference 1).

  In such a control device, a process of recognizing the position of a parking frame line (such as a white line drawn on a road surface) in a captured image of a camera and automatically or semi-automatically setting a target parking position based on this is performed. .

  However, since the process for setting the target parking position is based on the premise that the road surface is flat, in order to improve drainage, etc., when the road surface gradient is added within the parking frame line, the parking frame The vehicle may not be correctly guided within the line.

Here, an invention is disclosed for a vehicle upper viewpoint image composition device that combines images captured by a plurality of cameras that capture the periphery of a vehicle and displays a composite image when viewed from above the vehicle in a vehicle interior ( For example, see Patent Document 1). This apparatus is provided with means for detecting a change in road surface inclination, and the coordinate conversion from the image to the upward view is performed by reflecting the detection result. It is also considered that the above-mentioned problems can be solved by applying the technology described in the literature to the parking assistance device.
JP 2007-318460 A 4. Toyota Motor Corporation, “Prius New Car Guide (Part No. 7108100)”, Toyota Motor Corporation Service Department, issued on September 1, 2003. Electrical, p120-133

  However, in the above-mentioned Patent Document 1, as a specific example of a method for detecting a change in road surface inclination, (1) a method using altitude data in map data, (2) a method using a gyro sensor, and (3) an ultrasonic sensor are used. Methods and (4) a method of detecting an inclination by capturing a grid image projected on a road surface are disclosed, but it is not realistic to detect a road surface gradient in a parking frame line by these methods. The reason is described below.

  As for (1), it is extremely difficult to create map data including precise altitude data of any parking lot. For (2), only the road gradient around the vehicle is known, and the road gradient at a position away from the vehicle cannot be detected. For (3), only the road surface gradient at a position close to the vehicle is known due to the attenuation of ultrasonic waves, and the road surface gradient at a position away from the vehicle cannot be detected. In addition, since the road surface at a position away from the vehicle has a large inclination with respect to the ultrasonic sensor, reflection of ultrasonic waves cannot be expected so much. Regarding (4), it is necessary to mount a dedicated device for projecting a grid image on the road surface, which causes problems such as an increase in weight and cost, and a need for a new installation space.

  This invention is for solving such a subject, and provides the parking assistance apparatus which can guide a vehicle into a parking frame line more correctly, without increasing the size etc. of an apparatus. The main purpose.

In order to achieve the above object, one embodiment of the present invention provides:
Imaging means for imaging the rear of the vehicle;
In the image picked up by the image pickup means, on-image target parking position setting means for setting a target parking position in accordance with the picked-up parking frame line;
Target guidance position setting means for performing a predetermined conversion process on the target parking position set by the target parking position setting means on the image and determining a target guidance position;
A parking assistance device comprising guidance means for guiding the vehicle to the target guidance position determined by the target guidance position setting means,
The target guidance position setting unit corrects the target guidance position based on the position and shape of the parking frame line in the image captured by the imaging unit.
Parking assistance device.

  Note that “correcting the target guidance position” means correcting the target guidance position directly after performing the predetermined conversion process and correcting the target parking position before performing the predetermined conversion process. Includes those that indirectly correct the target guidance position.

  According to this aspect of the present invention, the target guidance position is corrected based on the position and shape of the parking frame line in the image captured by the imaging means, so that the vehicle can be more accurately detected without increasing the size of the device. It can be guided within the parking frame.

In one embodiment of the present invention,
For example, the target guidance position setting unit estimates a road surface gradient in the parking frame line based on the position and shape of the parking frame line in the image captured by the imaging unit, and based on the estimation result, the target of the vehicle It is a means for correcting the guidance position.

In one embodiment of the present invention,
Storage means for storing the position and shape of the parking frame line in the image captured by the imaging unit in association with the road surface gradient in the parking frame line;
The target guidance position setting unit is a unit that corrects the target guidance position by comparing the position and shape of the parking frame in the image captured by the imaging unit with the information stored in the storage unit. Also good.

  ADVANTAGE OF THE INVENTION According to this invention, the parking assistance apparatus which can guide a vehicle into a parking frame line more correctly can be provided, without increasing the size etc. of an apparatus.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the accompanying drawings.

  Hereinafter, the parking assistance apparatus 1 which concerns on one Example of this invention is demonstrated. FIG. 1 is a diagram illustrating an example of the overall configuration of a parking assistance apparatus 1 according to an embodiment of the present invention. The parking assistance device 1 is connected to a vehicle speed sensor 50, a shift position sensor 60, a height sensor 70, a steering device 80, a parking assistance ECU (Electronic Control Unit) 90, and a parking assistance ECU 90 connected to the multiplex communication line 99. The back monitor camera 10, the main switch 20, the display panel 30, and the speaker 40 are provided.

  The back monitor camera 10 includes an image sensor such as a charge-coupled device (CCD) and a complementary metal oxide semiconductor (CMOS) and a wide-angle lens, and is attached to the rear portion of the vehicle. The attachment angle and the angle of view (viewing angle) of the back monitor camera 10 are appropriately determined so that the scenery behind the vehicle is imaged at a predetermined angle approximately symmetrically about the vehicle longitudinal axis.

  The main switch 20 is, for example, a GUI (Graphical User Interface) switch set on a display image of the display panel 30 configured as a touch panel, and is used for inputting a vehicle guidance start instruction by the user. Not only this but the mechanical switch arrange | positioned in the vehicle interior may be sufficient. Further, as shown in FIG. 2, on the display image of the display panel 30, an arrow switch 22, a confirmation switch 24, and the like are set as GUI switches.

  The vehicle guidance in the present embodiment is, as will be described later, performing automatic steering control by the steering device 80 so that the vehicle can move backward to a target parking position set to be superimposed on the image captured by the back monitor camera 10. Say. At this time, the reverse driving force uses creep torque. However, the present invention is not limited to this, and the reverse driving force may be automatically output, or advice regarding the turning angle of the steering may be given or the parking guidance line may be displayed on the display panel 30. Note that the vehicle guidance may be selected from a plurality of control modes such as parallel parking control and garage entry control, but for the application of the present invention, such a difference in the control mode is not an essential problem. Don't do it.

  The vehicle speed sensor 50 includes, for example, a wheel speed sensor attached to each wheel and a skid control computer. The skid control computer converts a wheel speed pulse signal output from the wheel speed sensor into a vehicle speed rectangular wave pulse signal (vehicle speed signal). And output to the multiplex communication line 99. Shift position sensor, D (drive) for forward movement, 1, 2, R for reverse (reverse), P for parking (parking), N (neutral) for separating drive source and axle, etc. Is detected and output to the multiplex communication line 99. The height sensor 70 is attached to, for example, a rear cross member and detects a change in vehicle height.

  The steering device 80 includes, for example, a steering angle sensor 82, a torque sensor 84, an assist motor 86, and a controller 88 that controls the entire steering device. The steering angle sensor 82 is disposed inside the steering column, for example, and transmits a steering angle signal to the multiplex communication line 99 and the controller 88. The torque sensor 84 is disposed, for example, inside the steering column, and transmits a signal corresponding to the steering torque to the controller 88 by detecting torsion of a torsion bar attached between the input shaft and the output shaft. The assist motor 86 is, for example, a DC motor disposed in the column housing part, and assists the driver's steering operation by outputting torque necessary for steering the vehicle. The torque output from the assist motor 86 is deflected by the worm gear and the wheel gear, decelerated and transmitted to the column shaft, and finally changes the direction of the wheel. The controller 88 outputs a torque necessary for steering the vehicle based on a steering torque signal from the torque sensor 84 and other vehicle state signals (vehicle speed, yaw rate, etc.) during normal times when vehicle guidance is not performed. A control signal is output to the drive circuit of the assist motor 86. When the vehicle is being guided, the assist motor 86 is controlled based on an instruction signal from the parking assist ECU 90.

  The parking assistance ECU 90 is, for example, a computer unit in which a ROM, a RAM, and the like that store programs and the like are connected to each other via a bus, and a reading such as a hard disk, a DVD, a CD-ROM, an EEPROM, etc. A memory 91, an input / output interface, a timer, a counter, etc.

  As shown in FIG. 3, the parking support ECU 90 includes a master control unit 92, an on-image target parking position setting unit 93, and a correction processing unit as main functional blocks that function by executing a program stored in the ROM. 94, a coordinate conversion unit 95, and a vehicle guidance unit 96. Note that these functional blocks do not necessarily have to be based on another program, and a part for realizing a plurality of functional blocks may be included in the same program.

  The master control unit 92 monitors the output value of the shift position sensor 60 and displays the captured image of the back monitor camera 10 on the display panel 30 when the shift position is R (reverse).

  The master control unit 92 monitors the output signals of the vehicle speed sensor 50 and the main switch 20 in the initial state, and the output signal (vehicle speed signal) of the vehicle speed sensor 50 is zero (in reality, less than a minute value). When the output signal from the main switch 20 indicates an ON state, vehicle guidance is started. That is, vehicle guidance is started when the reverse shift position is selected, the vehicle is stopped, and the main switch 20 is operated.

  Vehicle guidance is performed in cooperation with other functional blocks. First, the master control unit 92 instructs the on-image target parking position setting unit 93 to recognize the parking frame line.

  The on-image target parking position setting unit 93 recognizes a parking frame line drawn on the road by a white line or the like in the captured image of the back monitor camera 40. The parking frame line is recognized by, for example, extracting a feature point (a point where the luminance change compared to the adjacent pixel is equal to or greater than a threshold value) in the captured image of the back monitor camera 40 and using a straight line extraction method such as a Hough transform. The feature points arranged in a straight line are recognized as the outline of the parking frame line. Then, pattern matching processing or the like is performed on the parking frame line, and for example, an image is taken of an inner region where the parking frame line is drawn in a “U” shape, a box shape, or two opposite sides. Set as the target parking position in the image.

  When the on-image target parking position setting unit 93 sets the target parking position by image recognition, the master control unit 92 causes the display panel 30 to display the target parking position superimposed on the captured image of the back monitor camera 10. FIG. 2 described above is an example of the display screen of the display panel 30 in this state. In this state, the user touches the arrow switch 22 as necessary to correct the target parking position and touches the confirmation switch 24 to confirm the target parking position in the captured image.

  The coordinate conversion unit 95 converts the target parking position in the captured image set by the on-image target parking position setting unit 93 into a target guidance position on a virtual two-dimensional plane viewed from above. The virtual two-dimensional plane is set in a predetermined area on the RAM with reference to the rear end of the vehicle, for example. Such conversion can be performed based on, for example, a conversion table 91A stored in the memory 91.

  The conversion table 91A is a table in which each pixel in the captured image of the back monitor camera 40 is associated with the coordinates on the virtual two-dimensional plane on a one-to-one basis, and the installation parameters (roll, pan, pitch, It is created in advance reflecting the installed height). In addition, the coordinate conversion unit 95 corrects the target guidance position according to the output of the height sensor 70. Such processing may be performed based on a predetermined correction algorithm, but a plurality of conversion tables may be provided in advance, and an appropriate one may be selected according to the output of the height sensor 70. FIG. 4 is a diagram illustrating a state in which the target parking position in the captured image is converted to the target guidance position on the virtual two-dimensional plane. In this figure, the virtual two-dimensional plane is expressed as an orthogonal coordinate system, but may be a polar coordinate system.

  The vehicle guidance unit 96 generates a route from the current position of the vehicle to the target guidance position, for example, by a combination of straight lines, clothoid curves, and arcs. Then, the target steering angle derived based on the curvature at each point on the route is output to the multiplex communication line 99. The controller 88 determines the steering torque of the assist motor 86 according to the output target steering angle. As a result, the vehicle moves to the target guidance position along a gentle curve, and it is possible to avoid the steering operation during parking, which is difficult and troublesome for a driver with low skill level. As described above, the reverse driving force required for the movement is such that the driver is instructed to turn off the accelerator and the engine creep torque is output. A driving force may be output.

  However, in an actual parking lot, there is a case where a road surface gradient is provided in the parking frame line in order to improve drainage performance. As a result, the target parking position is guided to a target on a virtual two-dimensional plane. It may not be possible to accurately convert to a position.

  FIG. 5 (A) shows the target parking position set in the captured image in the case where the inside of the parking frame line is an ascending road surface gradient (the back side of the parking frame is lower than the near side). 6 (A) shows the target parking position set in the captured image when the parking frame line has a downward road surface gradient (the back side of the parking frame is higher than the near side). In these cases, the position and shape of the parking frame line in the captured image are different, although the position of the parking frame line viewed from above is the same as the case where the parking frame line is flat, If the target parking position set in accordance with the parking frame line in the captured image is converted into the target guidance position, a deviation from the desired target guidance position is generated. 5 (B) and 6 (B) are diagrams showing target guidance positions set by converting the target parking positions shown in FIGS. 5 (A) and 6 (A). As a result of such a phenomenon, the vehicle may be guided to a position protruding from the parking frame line or stopped at a position inclined with respect to the parking frame line.

  Therefore, the parking assist device 1 of the present embodiment includes the correction processing unit 94 for correcting the target guidance position based on the position and shape of the parking frame line in the captured image. The correction processing unit 94 of the present embodiment estimates the road surface gradient in the parking frame line based on the position and shape of the parking frame line in the captured image, and corrects the vehicle guidance target based on the estimation result.

  More specifically, the road surface gradient in the parking frame line is changed by a predetermined road surface gradient (for example, zero (flat), ± 3 degrees, ± 6 degrees, ± 9 degrees,...) For a plurality of parking frame line positions. Information obtained as a result of imaging the position and shape of the parking frame line, that is, (A) the position of the parking frame line relative to the vehicle, and (B) the road surface gradient in the parking frame line as a data label. The position of the parking frame line on the image is stored in the memory 91 as correction data 91B. FIG. 7 is a diagram illustrating the content of the correction data 91B. Then, the correction processing unit 94 acquires the position and shape of the parking frame line in the actual captured image from the on-image target parking position setting unit 93, and searches for and reads the data closest thereto from the correction data 91B. Thereby, the road surface gradient indicated by the data label of the data retrieved from the correction data 91B can be estimated as the road surface gradient in the parking frame line.

  In addition, the position and shape of the parking frame line on the captured image are stored in the memory 91 as correction data 91B only for the positions of a plurality of parking frame lines when the inside of the parking frame line is flat. The road surface gradient in the parking frame line may be calculated based on the amount of deviation from this. Further, instead of the position and shape of the parking frame line in the actual captured image, data closest to the target parking position corrected by the user may be retrieved from the correction data 91B and read. This is because the target parking position corrected by the user is based on the position and shape of the parking frame line in the actual captured image.

  Then, the correction processing unit 94 reads, from the correction data 91B, data having the data label of the position of the parking frame line with respect to the vehicle in the data label searched and read from the correction data 91B and “road slope zero”. This is output to the coordinate conversion unit 95. In addition, when it is estimated that the road surface gradient in the parking frame line is zero, the correction processing unit 94 outputs only a signal indicating that to the coordinate conversion unit 95.

  In the coordinate conversion unit 95, for example, when a signal indicating that the road surface gradient in the parking frame line is zero is input, the target parking position set on the image is converted as described above to obtain the target guidance position. Set. On the other hand, when data representing the position of the parking frame line on the captured image is input, the position of the parking frame line included in the data input from the correction processing unit 94 is converted to set the target guidance position.

  With this process, even when there is a road gradient in the parking frame line, the target guidance position is set at the same position as when the parking frame line is flat. Therefore, it is possible to suppress the inconvenience shown in FIG. 5 and FIG.

  In the present invention, only a program corresponding to the correction data 91B and the correction processing unit 94 is newly required. In other words, since only the software configuration is added to the conventional parking assist device, problems such as an increase in the size and weight of the device and the need for a new installation space do not occur.

  According to the vehicle parking assistance apparatus 1 of the present embodiment, the vehicle can be guided more accurately into the parking frame line without increasing the size of the apparatus.

  The best mode for carrying out the present invention has been described above with reference to the embodiments. However, the present invention is not limited to these embodiments, and various modifications can be made without departing from the scope of the present invention. And substitutions can be added.

  For example, the configuration has been described in which the road surface gradient in the parking frame line is estimated and the target guidance position is corrected, but the process is further expanded to estimate the road surface gradient in the path from the vehicle to the parking frame line by the same principle. Then, the target guide position may be corrected.

  In addition, although correction based on the road surface gradient is performed as preprocessing for coordinate conversion, correction based on road surface gradient may be performed as post-processing for coordinate conversion. In this case, a road surface gradient may be estimated by the same processing as in the embodiment, and a table defining a correction amount for correcting the target guidance position based on the road surface gradient may be stored in the memory 91.

  The present invention can be used in the automobile manufacturing industry, the automobile parts manufacturing industry, and the like.

It is a figure which shows an example of the whole structure of the parking assistance apparatus 1 which concerns on one Example of this invention. 4 is a diagram illustrating an example of a display image on the display panel 30. FIG. It is a figure which shows the main functional blocks which ECU90 for parking assistance has. It is a figure which shows a mode that the target parking position in a captured image is converted into the target guidance position on a virtual two-dimensional plane. It is a figure which shows the target parking position set when there is a road surface gradient in a parking frame line, and the target guidance position when this is converted. It is a figure which shows the target parking position set when there is a road surface gradient in a parking frame line, and the target guidance position when this is converted. It is the figure which illustrated the contents of data for correction 91B.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Parking assistance apparatus 10 Back monitor camera 20 Main switch 22 Arrow switch 24 Confirmation switch 30 Display panel 40 Speaker 50 Vehicle speed sensor 60 Shift position sensor 70 Height sensor 80 Steering device 82 Steering angle sensor 84 Torque sensor 86 Assist motor 88 Controller 90 Parking assistance ECU
91 Memory 91A Coordinate conversion table 91B Correction data 92 Master control unit 93 Image target parking position setting unit 94 Correction processing unit 95 Coordinate conversion unit 96 Vehicle guidance unit 99 Multiplex communication line

Claims (3)

Imaging means for imaging the rear of the vehicle;
In the image picked up by the image pickup means, on-image target parking position setting means for setting a target parking position in accordance with the picked-up parking frame line;
Target guidance position setting means for performing a predetermined conversion process on the target parking position set by the target parking position setting means on the image and determining a target guidance position;
A parking assistance device comprising guidance means for guiding the vehicle to the target guidance position determined by the target guidance position setting means,
The target guidance position setting unit corrects the target guidance position based on the position and shape of the parking frame line in the image captured by the imaging unit.
Parking assistance device. The target guidance position setting unit estimates a road surface gradient in the parking frame line based on the position and shape of the parking frame line in the image captured by the imaging unit, and based on the estimation result, the target guidance position of the vehicle Is a means of correcting
The parking assistance device according to claim 1. Storage means for storing the position and shape of the parking frame line in the image captured by the imaging unit in association with the road surface gradient in the parking frame line;
The target guidance position setting unit is a unit that corrects the target guidance position by comparing the position and shape of the parking frame in the image captured by the imaging unit with information stored in the storage unit.
The parking assistance device according to claim 1 or 2.

Images