Patent Application
20100066515
The present invention relates to a parking assistance apparatus and, in particular, a parking assistance apparatus that performs a parking assist by recognizing a relative positional relation between a vehicle and a target parking position with reliability.
In addition, the present invention also relates to a parking assistance apparatus part that realizes such a parking assistance apparatus through connection to a camera, a parking assist method for performing such a parking assist, and a parking assist program for causing a computer to execute the method.
Still in addition, the present invention also relates to a method of calculating a vehicle travel parameter such as a turn radius with respect to a steering angle, a vehicle travel parameter calculation program for causing a computer to execute this method, a vehicle travel parameter calculation apparatus, and a vehicle travel parameter calculation apparatus part.
Conventionally, for instance, as disclosed in Patent Document 1, a parking assisting device has been developed which takes an image behind a vehicle with a CCD camera, recognizes a parking zone behind the vehicle from the obtained image, calculates a target parking path from a current stop position of the vehicle to the parting zone, and gives a certain steering angle corresponding to this target parking path to a driver. When the driver drives the vehicle backward while constantly maintaining a steering angle at the given value and temporarily stops the vehicle at a location at which the steering angle should be changed, a new target parking path from there to the parking zone is calculated and a certain steering angle corresponding to this new target parking path is given to the driver again. The driver can drive the vehicle into the parking zone that is a target by driving the vehicle backward while constantly maintaining the steering angle at the newly given value.
When such a parking assist is performed, in order to move the vehicle along the target parking path, it is required to grasp a current state of the vehicle. For instance, in Patent Document 2, an apparatus is disclosed which takes an image in front of or behind a vehicle, extracts information about lightness in a predetermined area that is horizontal to a road surface, and detects a yaw rate of the vehicle based on a lightness gradient and a time-varying degree of this lightness information. When such an apparatus is used, it becomes possible to grasp a yaw rate of a vehicle from image information.
[Patent Document 1] JP 2002-172988 A
[Patent Document 2] JP 04-151562 A
In the device disclosed in Patent Document 1, however, it is tried to improve parking accuracy by newly calculating a target parking path when a vehicle is temporarily stopped at a changing point of a steering angle but it is difficult to accurately identify a relative positional relation between a parking zone and a current position of the vehicle merely by recognizing the parking zone behind the vehicle from an image taken by a CCD camera. This leads to a problem that although the target parking path is recalculated at the changing point of the steering angle, it is difficult to complete parking with high accuracy.
Also, in the apparatus disclosed in Patent Document 2, it is possible to detect a yaw rate of a vehicle but it is impossible to calculate a vehicle travel parameter such as a turn radius with respect to a steering angle, which results in a problem that an enormous amount of labor and time is required to obtain this vehicle travel parameter through actual measurement. In addition, there is also a problem that the vehicle travel parameter obtained through actual measurement includes an error due to various factors.
The present invention has been made in light of such conventional problems, and has an object to provide a parking assistance apparatus with which it becomes possible to park a vehicle at a target parking position with accuracy.
In addition, the present invention has an object to provide a parking assistance apparatus part that realizes such a parking assistance apparatus through connection to a camera, a parking assist method for performing such a parking assist, and a parking assist program for causing a computer to execute the method.
Further, the present invention has an object to provide a vehicle travel parameter calculation method with which it becomes possible to obtain a vehicle travel parameter with ease and accuracy, a vehicle travel parameter calculation program for causing a computer to execute such a calculation method, a vehicle travel parameter calculation apparatus, and a vehicle travel parameter calculation apparatus part.
A parking assistance apparatus according to the preset invention includes: a camera mounted on a vehicle for taking an image of a fixed target that is fixed to a predetermined place having a predetermined positional relation with respect to a target parking position and has at least one characteristic point; image processing means for extracting the characteristic point of the fixed target based on the image of the fixed target taken by the camera and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; positional parameter calculation means for calculating positional parameters of the camera including at least two-dimensional coordinates and a pan angle with reference to the fixed target based on two or more sets of the two-dimensional coordinates recognized by the image processing means; relative position identification means for identifying a relative positional relation between the vehicle and the target parking position based on the positional parameters of the camera calculated by the positional parameter calculation means and the predetermined positional relation of the fixed target with respect to the target parking position; and parking locus calculation means for calculating a parking locus for leading the vehicle to the target parking position based on the relative positional relation between the vehicle and the target parking position identified by the relative position identification means.
A parking assistance apparatus part according to the present invention includes: an input portion connected to a camera mounted on a vehicle for taking an image of a fixed target that is fixed to a predetermined place having a predetermined positional relation with respect to a target parking position and has at least one characteristic point; image processing means for extracting the characteristic point of the fixed target based on the image of the fixed target taken by the camera and inputted through the input portion and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; positional parameter calculation means for calculating positional parameters of the camera including at least two-dimensional coordinates and a pan angle with reference to the fixed target based on two or more sets of the two-dimensional coordinates recognized by the image processing means; relative position identification means for identifying a relative positional relation between the vehicle and the target parking position based on the positional parameters of the camera calculated by the positional parameter calculation means and the predetermined positional relation of the fixed target with respect to the target parking position; and parking locus calculation means for calculating a parking locus for leading the vehicle to the target parking position based on the relative positional relation between the vehicle and the target parking position identified by the relative position identification means.
A parking assist method according to the present invention includes the steps of: taking an image of a fixed target, which is fixed to a predetermined place having a predetermined positional relation with respect to a target parking position and has at least one characteristic point, with a camera mounted on a vehicle; extracting the characteristic point of the fixed target based on the taken image of the fixed target and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; calculating positional parameters of the camera including at least two-dimensional coordinates and a pan angle with reference to the fixed target based on two or more sets of the recognized two-dimensional coordinates; identifying a relative positional relation between the vehicle and the target parking position based on the calculated positional parameters of the camera and the predetermined positional relation of the fixed target with respect to the target parking position; and calculating a parking locus for leading the vehicle to the target parking position based on the identified relative positional relation between the vehicle and the target parking position.
A parking assist program according to the present invention causes a computer to execute the steps of: taking an image of a fixed target, which is fixed to a predetermined place having a predetermined positional relation with respect to a target parking position and has at least one characteristic point, with a camera mounted on a vehicle; extracting the characteristic point of the fixed target based on the taken image of the fixed target and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; calculating positional parameters of the camera including at least two-dimensional coordinates and a pan angle with reference to the fixed target based on two or more sets of the recognized two-dimensional coordinates; identifying a relative positional relation between the vehicle and the target parking position based on the calculated positional parameters of the camera and the predetermined positional relation of the fixed target with respect to the target parking position; and calculating a parking locus for leading the vehicle to the target parking position based on the identified relative positional relation between the vehicle and the target parking position.
A vehicle travel parameter calculation method according to the present invention includes the steps of: causing a vehicle to travel; capturing a detection signal from a sensor concerning vehicle travel; taking an image of a fixed target being outside the vehicle and having a characteristic point with a camera mounted on the vehicle at each of two locations midway through the travel; extracting the characteristic point of the fixed target for each taken image of the fixed target and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; calculating each of positional parameters of the camera including two-dimensional coordinates and a pan angle with reference to the fixed target at the two locations based on the recognized two-dimensional coordinates; and calculating a travel parameter of the vehicle based on at least two sets of the calculated positional parameters and the captured detection signal.
A vehicle travel parameter calculation program according to the present invention causes a computer to execute the steps of: capturing a detection signal from a sensor concerning vehicle travel at a time of travel of a vehicle; taking an image of a fixed target being outside the vehicle and having a characteristic point with a camera mounted on the vehicle at each of at least two locations midway through the travel; extracting the characteristic point of the fixed target for each taken image of the fixed target and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; calculating each of positional parameters of the camera including two-dimensional coordinates and a pan angle with reference to the fixed target at the at least two locations based on the recognized two-dimensional coordinates; and calculating a travel parameter of the vehicle based on at least two sets of the calculated positional parameters and the captured detection signal.
A vehicle travel parameter calculation apparatus according to the present invention includes: a sensor for obtaining a detection signal concerning vehicle travel; a camera mounted on a vehicle for taking an image of a fixed target being outside the vehicle and having a characteristic point; image processing means for extracting the characteristic point of the fixed target for each image of the fixed target taken by the camera at least two locations midway through travel of the vehicle and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; positional parameter calculation means for calculating each of positional parameters of the camera including two-dimensional coordinates and a pan angle with reference to the fixed target at the at least two locations based on the two-dimensional coordinates recognized by the image processing means; and vehicle travel parameter calculation means for calculating a travel parameter of the vehicle based on at least two sets of the positional parameters calculated by the positional parameter calculation means and the detection signal obtained by the sensor.
Further, a vehicle travel parameter calculation apparatus part according to the present invention includes: an input portion connected to a camera mounted on a vehicle for taking an image of a fixed target being outside the vehicle and having a characteristic point; image processing means for extracting the characteristic point of the fixed target for each image of the fixed target taken by the camera at least two locations midway through travel of the vehicle and inputted through the input portion and recognizing two-dimensional coordinates of the characteristic point on the image of the fixed target; positional parameter calculation means for calculating each of positional parameters of the camera including two-dimensional coordinates and a pan angle with reference to the fixed target at the at least two locations based on the two-dimensional coordinates recognized by the image processing means; and vehicle travel parameter calculation means, which is connected to a sensor that obtains a detection signal concerning vehicle travel, for calculating a travel parameter of the vehicle based on at least two sets of the positional parameters calculated by the positional parameter calculation means and the detection signal obtained by the sensor.
According to the present invention, it becomes possible to park a vehicle at a target parking position with accuracy by identifying a relative positional relation between the vehicle and the target parking position.
Also, according to the present invention, it becomes possible to obtain a vehicle travel parameter with ease and accuracy.
Hereinafter, embodiments of the present invention will be described based on the accompanying drawings.
A construction of a parking assistance apparatus according to a first embodiment of the present invention is shown in
As shown in
Also, the mark M is fixed at a predetermined place having a predetermined positional relation with respect to the parking space S and it is assumed that the predetermined positional relation of the mark M with respect to the parking space S is grasped in advance. As this mark M, for instance, as shown in
Next, an operation of the first embodiment will be described with reference to a flowchart of
First, in Step S1, in a state in which, as shown in
The image taken by the camera 1 is inputted into the image processing means 2 through the input portion K and, in subsequent Step S2, the image processing means 2 extracts the five characteristic points C1 to C5 of the mark M from the image of the mark M taken by the camera 1 and recognizes and obtains each of two-dimensional coordinates of those characteristic points C1 to C5 on the image.
Next, in Step S3, based on the two-dimensional coordinates of each of the characteristic points C1 to C5 recognized by the image processing means 2, the positional parameter calculation means 3 calculates positional parameters including six parameters that are three-dimensional coordinates (x, y, z), a tilt angle (dip angle), a pan angle (direction angle) and a swing angle (rotation angle) of the camera 1 with reference to the mark M.
Here, a positional parameter calculation method by the positional parameter calculation means 3 will be described.
First, a point on the ground dropped from a center of a rear axle of the vehicle 7 vertically with respect to a road surface is set as an origin O, a road surface coordinate system is assumed in which an x axis and a y axis are set in a horizontal direction and a z axis is set in a vertical direction, and an image coordinate system is assumed in which an X axis and a Y axis are set on the image taken by the camera 1.
Coordinate values Xm and Ym (m=1 to 5) of the characteristic points C1 to C5 of the mark M in the image coordinate system are expressed by the following expressions from the six positional parameters of the characteristic points C1 to C5 of the mark M in the road surface coordinate system, in other words, coordinate values xm, ym and zm, and angle parameters Kn (n=1 to 3) that are the tilt angle (dip angle), the pan angle (direction angle) and the swing angle (rotation angle) described above using functions F and G.
Xm=F(xm,ym,zm,Kn)+DXm
Ym=G(xm,ym,zm,Kn)+DYm
Here, DXm and DYm are deviations between the X coordinates and the Y coordinates of the characteristic points C1 to C5 calculated using the functions F and G, and the coordinate values Xm and Ym of the characteristic points C1 to C5 recognized by the image processing means 2.
In other words, through expression of each of the X coordinates and the Y coordinates of the five characteristic points C1 to C5, ten relational expressions are created in total with respect to the six positional parameters (xm, ym, zm, Kn).
Therefore, the positional parameters (xm, ym, zm, Kn) are obtained which minimizes the following sum of squares of the deviations DXm and DYm.
S=Σ(DXm2+DYm2)
In other words, an optimization problem that minimizes S is solved. It is possible to use a known optimization method such as a simplex method, a steepest descent method, a Newton method, or a quasi-Newton method.
It should be noted here that by creating relational expressions whose number is greater than the number “six” of the positional parameters (xm, ym, zm, Kn) to be calculated, the positional parameters are determined, so it becomes possible to obtain the positional parameters (xm, ym, zm, Kn) with accuracy.
In the first embodiment of the present invention, ten relational expressions are created for the six positional parameters (xm, ym, zm, Kn) from the five characteristic points C1 to C5, but it is sufficient that the number of the relational expressions is equal to or greater than the number of the positional parameters (xm, ym, zm, Kn) to be calculated and when six relational expressions are created from at least three characteristic points, it is possible to calculate six positional parameters (xm, ym, zm, Kn).
In Step S4, using the positional parameters of the camera 1 thus calculated, the relative position identification means 4 identifies a relative positional relation between the vehicle 7 and the parking space S. In other words, the relative positional relation between the camera 1 and the parking space S is identified based on the positional parameters calculated by the positional parameter calculation means 3 and the predetermined positional relation of the mark M with respect to the parking space S grasped in advance and, further, the relative positional relation between the vehicle 7 and the parking space S is identified because the predetermined positional relation of the camera 1 with respect to the vehicle 7 is grasped in advance.
Next, in Step S5, the parking locus calculation means 5 calculates a parking locus for leading the vehicle 7 into the parking space S based on the relative positional relation between the vehicle 7 and the parking space S identified by the relative position identification means 4.
For instance, as shown in
It should be noted here that it is possible to construct the image processing means 2, the positional parameter calculation means 3, the relative position identification means 4 and the parking locus calculation means 5 from a computer, and by setting a parking assist program of the operations in Steps S1 to S5 of
Also, it is possible to collectively form the parking assistance apparatus part P1 constructed by the input portion K, the image processing means 2, the positional parameter calculation means 3, the relative position identification means 4 and the parking locus calculation means 5 in a form of a substrate module, a chip, or the like and a parking assistance apparatus is realized merely by connecting the camera 1 mounted on the vehicle to the input portion K of this parking assistance apparatus part P1. Further, when the guide apparatus 6 is connected to the parking locus calculation means 5 of the parking assistance apparatus part P1, it becomes possible to output the drive operation guide information described above to the driver of the vehicle 7.
It should be noted here that in the first embodiment described above, the positional parameters including the six parameters that are the three-dimensional coordinates (x, y, z), the tilt angle (dip angle), the pan angle (direction angle) and the swing angle (rotation angle) of the camera 1 with reference to the mark M are calculated, so even when there exists a step or an inclination between the floor surface of the parking space S, in which the mark M is arranged, and the road surface at a current position of the vehicle 7, it becomes possible to perform a highly accurate parking assist by correctly identifying the relative positional relation between the mark M and the vehicle 7.
However, when there exists no inclination between the floor surface of the parking space S, in which the mark M is arranged, and the road surface at the current position of the vehicle 7, it is possible to identify the relative positional relation between the mark M and the vehicle 7 by calculating positional parameters including at least four parameters that are the three-dimensional coordinates (x, y, z) and the pan angle (direction angle) of the camera 1 with reference to the mark M. In this case, when four relational expressions are created from two-dimensional coordinates of at least two characteristic points of the mark M, it is possible to obtain four positional parameters but it is preferable that the four positional parameters be accurately calculated using two-dimensional coordinates of more characteristic points with a least square method or the like.
In addition, when the mark M and the vehicle 7 exist on the same plane and there exists no step and inclination between the floor surface of the parking space S, in which the mark M is arranged, and the road surface at the current position of the vehicle 7, when positional parameters including at least three parameters that are two-dimensional coordinates (x, y) and the pan angle (direction angle) of the camera 1 with reference to the mark M are calculated, it is possible to identify the relative positional relation between the mark M and the vehicle 7. Also in this case, when four relational expressions are created from two-dimensional coordinates of at least two characteristic points of the mark M, it is possible to obtain three positional parameters but it is preferable that the three positional parameters be accurately calculated using two-dimensional coordinates of more characteristic points with a least square method or the like.
In the first embodiment, the camera 1 is embedded in the door mirror 8 positioned in a side portion of the vehicle 7, but as shown in
In the first and second embodiments described above, a case where lateral parking into the parking space S is performed has been described as an example. In a like manner, it is also possible to, as shown in
In addition, it is also possible to perform the parallel parking by, as shown in
It should be noted that, it is required to instruct the parking locus calculation means 5 which one of the lateral parking and the parallel parking is to be performed. A construction is also possible in which a selection switch for selection of any one of a lateral mode and a parallel mode is provided near a driver's seat and the driver operates the selection switch. Alternatively, a construction is also possible in which a mark installed in the parking space for the lateral parking and a mark installed in the parking space for the parallel parking are set different from each other, the image processing means 2 distinguishes between the mark for the lateral parking and the mark for the parallel parking, and any one of the lateral parking and the parallel parking is automatically selected.
In the first to third embodiments described above, after an image of the mark M in the parking space S is taken by the camera 1 and the parking locus L is calculated, the vehicle 7 is led to the parking space S along the parking locus L, but a construction is also possible in which, as shown in
In other words, after the vehicle 7 is moved from the location C in accordance with the parking locus L first calculated by the parking locus calculation means 5, an image of the mark M in the parking space S is taken by the camera 1 again at a location F, new two-dimensional coordinates of the characteristic points C1 to C5 of the mark M are recognized by the image processing means 2, new positional parameters of the camera 1 are calculated by the positional parameter calculation means 3, a new relative positional relation between the vehicle 7 and the parking space S is identified by the relative position identification means 4, and a new parking locus L′ is calculated by the parking locus calculation means 5. Then, drive operation guide information for traveling the vehicle 7 along this new parking locus L′ is outputted from the guide apparatus 6 to the driver.
As the distance between the vehicle 7 and the parking space S is reduced, it becomes possible to recognize the mark M at a closer distance in a larger size, which improves a resolution with respect to the characteristic points C1 to C5 of the mark M and increases distances between the characteristic points C1 to C5. Therefore, it becomes possible to identify the relative positional relation between the mark M and the vehicle 7 with higher accuracy. As a result, by recalculating the new parking locus L′ in a state in which the distance between the vehicle 7 and the parking space S is reduced, it becomes possible to perform parking with a higher degree of accuracy.
In addition, a construction is possible in which the parking locus is recalculated from moment to moment at predetermined time intervals or moving distance intervals. With this construction, it becomes possible to perform the parking into the parking space S that is a final target parking position with accuracy with almost no influence by an error in initial recognition of the characteristic points C1 to C5 of the mark M, states of the vehicle 7 such as a worn condition of a tire and an inclination of the vehicle 7, states of a road surface such as a step and a tilt, or the like.
In
In the fourth embodiment, the new parking locus is recalculated in the state in which the distance between the vehicle 7 and the parking space S is reduced but it is also possible to obtain the new parking locus using a previously calculated parking locus.
For instance, in a case where parallel parking is performed, when, as shown in
When, as shown in
When a new parking locus L′ including the locus portions La′ and Lb′ is obtained in this manner, it becomes possible to reduce a load of the recalculation.
It should be noted here that also in lateral parking, when the parking locus is formed by combining multiple locus portions in a curve shape and locus portions in a straight line shape with each other, it is possible to apply the fifth embodiment in a like manner.
In the first to fifth embodiments described above, an image of the mark M in the parking space S is taken using the camera 1 installed in any one of the side portion and the rear portion of the vehicle 7, but it is also possible to, as shown in
For instance, when an image of the mark M in the parking space S is taken by the camera 1 in the side portion of the vehicle 7 at the location A in the vicinity of the parking space S, it becomes possible to recognize the mark M at a closer distance in a larger size, which makes it possible to identify the relative positional relation between the mark M and the vehicle 7 with higher accuracy and calculate an accurate parking locus L. Then, after the mark M in the parking space S enters into the field of view of the camera 9 in the rear portion of the vehicle 7, it is possible to take an image of the mark M with the camera 9. For instance, when the vehicle 7 is moved to a turn location B of
As a result, it becomes possible to perform a further accurate parking assist.
In the first to sixth embodiments described above, a figure having an external form in a square shape, in which four isosceles right-angled triangles are abutted against each other, is used as the mark M in the parking space S, but the present invention is not limited thereto. For example, it is possible to use various marks as shown in
A mark M1 shown in
A mark M2 shown in
A mark M3 shown in
It is possible to use those marks M2 and M3 in the same manner as the mark M of
For instance, when, as shown in
In addition, a mark M4 shown in
In the first to seventh embodiments described above, the mark in the parking space S has three or more characteristic points and by taking an image of the mark with the camera 1 or 9 at one location, six or more relational expressions are created and six positional parameters (xm, ym, zm, Kn) of the camera 1 or 9 are calculated, but it is also possible to use a mark that has only one or two characteristic points. Note that it is assumed that the vehicle 7 is provided with moving amount sensors for detecting a moving distance and a moving direction of the vehicle 7, such as a wheel speed sensor, a yaw rate sensor, and a GPS.
For instance, it is assumed that, as shown in
Next, the vehicle 7 is moved to a location A2. In this case, it is required that the location A2 is within a range in which the mark M5 is captured in the field of view of the camera 1. Further, a moving distance and a moving direction of the vehicle 7 from the location A1 to the location A2 are detected by the moving amount sensors provided to the vehicle 7. By taking an image of the mark M5 again with the camera 1 at the location A2, further four relational expressions expressing the X coordinates and Y coordinates in the image coordinate system of the two characteristic points C1 and C2 are obtained. Based on eight relational expressions including the four relational expressions obtained at the location A1 and the four relational expressions obtained at the location A2 and the relative positions of the location A1 and the location A2 detected by the moving amount sensors, it is possible to calculate six positional parameters (xm, ym, zm, Kn) of the camera 1. As a result, it becomes possible to identify the relative positional relation between the vehicle 7 and the parking space S and calculate the parking locus.
In a like manner, when a mark having only one characteristic point is installed in the parking space S, by taking an image of the mark at each of at least three locations, it becomes possible to obtain six or more relational expressions and calculate six positional parameters (xm, ym, zm, Kn) of the camera 1.
In addition, it is also possible to mount multiple cameras, at least part of whose fields of view overlap each other, on the vehicle 7, and simultaneously take images of the mark with those multiple cameras. For instance, when an image of a mark having two characteristic points is taken with each of two cameras, information that is equal to that in the case where an image of a mark having four characteristic points is taken with one camera is obtained. Also, when an image of a mark having one characteristic point is taken with each of three cameras, information that is equal to that in the case where an image of a mark having three characteristic points is taken with one camera is obtained. Accordingly, it becomes possible to calculate six positional parameters (xm, ym, zm, Kn).
By reducing the number of characteristic points in this manner, it becomes possible to reduce a size of the mark.
In the first to eighth embodiments described above, when the mark is installed on an entrance side of the parking space S in the case of lateral parking, a distance between the vehicle 7 positioned in the vicinity of the parking space S and the mark is reduced, so it becomes easy to perceive the mark and recognize its characteristic points. Note that, it is not necessarily required to install the mark on the entrance side of the parking space S. It is also possible to, as shown in
Also, as shown in
Further, as shown in
It should be noted here that it is preferable that the mark used in the present invention have a specific shape, color, and the like that are easy to discriminate with respect to a shape existing in the natural world, be a mark whose existence is easy to perceive through image recognition by the image processing means 2, and further, be a mark whose internally included characteristic points are easy to recognize.
Also, it is hoped that the mark has a sufficient size and is installed at a place, at which perception from the vehicle 7 is easy, so that target parking accuracy can be realized by accuracy of a relative positional relation between the vehicle 7 and the mark calculated based on the two-dimensional coordinates of the recognized characteristic points and accuracy of the parking locus calculated based on the relative positional relation.
More specifically, it is possible to install the mark by, for instance, directly painting it at a predetermined place such as a floor surface or a wall surface of the parking space S, sticking a sheet, on which the mark is drawn, at a predetermined place, or the like.
By displaying the mark in a form of a QR code or displaying the mark in a form of a two-dimensional barcode that is calibrated on a line parallel to or vertically to a side of the parking space S or on a diagonal line of the parking space S, it also becomes possible to store, in the mark, various information given below such as information concerning the parking space S itself and/or information concerning a method of parking into the parking space S, in addition to the characteristic points, and read the information through image recognition by the image processing means 2.
(1) Characteristics of the parking space S itself (such as a size, an inclination, deformation and a tilt)
(2) An address of the parking space S, a frame number in a large parking lot
In a large parking lot, a frame number is designated at an entrance and there is a case where a moving path in the parking lot is also guided. By identifying the frame number stored in the mark, it becomes possible for the vehicle to recognize which frame is a designated frame. Also, through cooperation with a navigation system, confirmation of a private garage and confirmation of an address of a garage at a destination become possible.
(3) A parking fee
(4) A parking use limitation (such as an available time zone, eligibility, and the presence or absence of a use right due to exclusive use by disabled persons or the like)
(5) A reachable range on the periphery of the parking lot, an entering limit range, the presence or absence and a position of an obstacle, and a condition at the time of parking (such as designation of forward parking)
Also, instead of the mark, a signboard may be set up at a predetermined place having a predetermined positional relation with respect to the parking space S, the various information described above may be displayed on this signboard, and the information may be read through image recognition by the image processing means 2.
In the first to ninth embodiments described above, it is also possible to display the mark used as a fixed target by light. For instance, as shown in
For instance, as shown in
When the mark M is displayed using light like in this tenth embodiment, a risk that a shape of the mark will be impaired due to a stain or rubbing of a mark installation surface is reduced as compared with a case where the mark is displayed through painting or using a sheet, which makes it possible to, even when the mark M is used for a long time, detect the relative positional relation between the vehicle 7 and the mark M with accuracy.
Also, by controlling the optical display apparatus 18 with the display control apparatus 19, it becomes possible to change display light intensity of the mark M with ease. Therefore, by adjusting the light intensity in accordance with brightness of a peripheral atmosphere such as in daytime or nighttime, it becomes possible to display the mark M that is easy to recognize at all times.
When the projector 20 or the laser scanner 21 is used as the optical display apparatus 18, by controlling the optical display apparatus 18 with the display control apparatus 19, it becomes possible to change a size of the mark M to be displayed with ease. Therefore, by displaying a large mark M when a distance of the vehicle 7 from the mark M is long and displaying a small mark M when the distance of the vehicle 7 from the mark M is reduced, recognition accuracy of the characteristic points of the mark M is improved. Note that in this case, it is required to transmit information concerning the size of the mark M to the vehicle 7 side.
In a like manner, when the projector 20 or the laser scanner 21 is used as the optical display apparatus 18, by controlling the optical display apparatus 18 with the display control apparatus 19, it becomes possible to change a position of the mark M to be displayed with ease. Therefore, when it is desired to adjust the target parking position in accordance with the presence of an obstacle in the parking space S or the like, it becomes possible to park the vehicle 7 at a desired position by changing the position of the mark M with ease. In addition, instead of installing multiple marks on the floor surface in the vicinity of the entrance of the parking space S and the floor surface of the back portion or the like as shown in
It should be noted here that also when the position of the mark M is changed in this manner, it is required to transmit information concerning the position of the mark M to the vehicle 7 side.
Even when the light-emitting apparatus 23 in a form of an electronic bulletin board shown in
When the projector 20 or the laser scanner 21 is used, it becomes possible to change a display color of the mark M with ease. Therefore, by adjusting the display color in accordance with a change of a peripheral atmosphere, it also becomes possible to display the mark M that is easy to recognize at all times.
Also, when the projector 20 or the laser scanner 21 is used, the mark M may be displayed on a plane like a screen installed on a floor surface, a side wall, or the like of the parking space S. In this case, even when the floor surface, the side wall, or the like of the parking space S includes projections and depressions, it becomes possible to display the mark M with no impairment of a mark shape, which improves recognition accuracy of the characteristic points of the mark M. Note that it is possible to realize the plane like a screen by selecting a material and a shape in accordance with an installation place through sticking of a flexible screen onto an installation surface, installation of a flat plate member, or the like.
It is also possible to modulate brightness, wavelength (color), or the like of display light of the mark M by controlling the optical display apparatus 18 with the display control apparatus 19 and demodulate an image of the mark M taken by the camera of the vehicle 7. In this case, it becomes possible to recognize positions of the characteristic points of the mark M with accuracy by excluding an influence of noise due to sunlight, illumination light, or the like. Also, through modulation of the display light of the mark M, it becomes possible to superimpose the various information described in the ninth embodiment, such as the information concerning the parking space S itself and/or the information concerning the method of parking into the parking space S, as well as the characteristic points on the mark M. For instance, it also becomes possible to superimpose information indicating that the mark M is a passage point to the target parking position or information indicating that the mark M is a parking completion position, while changing a display position of the mark M in accordance with a position of the vehicle 7.
It should be noted here that it is sufficient that the display light of the mark M is recognizable by the camera of the vehicle 7 and it is also possible to use non-visible light such as infrared rays or ultraviolet rays. In addition, it is also possible to use high-speed modulated display light unrecognizable with an ordinary human eye and, still in addition, it is also possible to perform so-called imprinting of the mark M into an image recognizable with a human eye by displaying the mark M in a very short time which is impossible to recognize with the human eye. By recognizing the mark M imprinted in this manner with the camera of the vehicle 7, the relative positional relation between the vehicle 7 and the mark M is detected. In a like manner, it is also possible to imprint the various information described above in an image or the mark M.
In the first to tenth embodiments described above, it is also possible to store a relative positional relation of a parking completion position with respect to the mark when the vehicle 7 is actually parked in accordance with the parking locus L calculated by the parking locus calculation means 5. And, the parking locus calculation means 5 may calculate a parking locus, which is corrected so that the vehicle 7 will be led to a target parking position, based on the stored relative positional relation between the previous parking completion position and the mark at the time of parking locus L calculation in the next parking operation.
In this case, when there occurs a positional deviation between a target parking position and an actual parking completion position, it becomes possible to compensate for the positional deviation. In addition, in the case of a private garage or the like, it also becomes possible to set not a center of a parking space but an eccentric place as a target parking position.
It should be noted here that it is possible to recognize the relative positional relation between the parking completion position and the mark by, for instance, as shown in
Further, a construction is also possible in which a navigation system is linked and when a specific parking space such as a private garage is perceived by the navigation system, a parking locus corrected based on a stored relative positional relation between a previous parking completion position and the mark is calculated. In this manner, it becomes possible to park the vehicle at a prescribed position in the case of an ordinary parking lot, and park the vehicle in a specially set condition such as at a position displaced from a center in the case of a specific parking space such as a private garage. Note that a GPS sensor may be provided instead of the navigation system to perceive a specific parking space based on information from the GPS sensor.
In the first to eleventh embodiments described above, it is possible to, as shown in
The guide information creation means 10 is means for creating drive operation guide information for traveling the vehicle 7 along the parking locus L based on detection signals from sensors concerning vehicle travel, such as a steering angle sensor 12, a yaw rate sensor 13, and a speed sensor 14, and the parking locus L calculated by the parking locus calculation means 5, and can be constructed from a computer.
The guide information output means 11 is means for outputting the guide information created by the guide information creation means 10 and can be constructed from, for instance, a speaker or a buzzer that transmits the guide information by stimulating the sense of hearing of the driver through emission of a voice, a warning sound, or the like. Aside from this, a display or a lamp that transmits the guide information by stimulating the sense of sight through image displaying, light emission, or the like may be used as the guide information output means 11. In addition, it is also possible to use a vibrator or the like, which transmits the guide information by stimulating the sense of touch through vibration or the like, as the guide information output means 11.
The guide information creation means 10 repeatedly captures a steering angle signal from the steering angle sensor 12, a yaw rate signal from the yaw rate sensor 13 and a speed pulse signal from the speed sensor 14 in accordance with travel of the vehicle 7 and calculates a turn radius, a turn angle and a moving distance of the vehicle 7 based on those signals. With this construction, a positional change amount from the relative positions of the vehicle 7 and the parking space S identified by the relative position identification means 4 in Step S4 of
The drive operation guide information created in this manner is outputted from the guide information output means 11 to the driver of the vehicle 7.
A construction of an apparatus for implementing a vehicle travel parameter calculation method according to a thirteenth embodiment of the present invention is shown in
It should be noted that it is assumed that in this thirteenth embodiment, a turn radius R with respect to a steering angle, a gain of the yaw rate sensor 13, and a moving distance per speed pulse are calculated as travel parameters of the vehicle.
The mark M installed on the road surface is the same as that used in the first embodiment and it is possible to, as shown in
Next, the vehicle travel parameter calculation method according to the thirteenth embodiment will be described with reference to a flowchart of
First, in Step S11, as shown in
The image taken by the camera 1 is inputted into the image processing means 2 through the input portion K and, in subsequent Step S12, the image processing means 2 extracts the five characteristic points C1 to C5 of the mark M from the image of the mark M taken by the camera 1 to recognize and obtain each of two-dimensional coordinates of those characteristic points C1 to C5 on the image.
Next, in Step S13, based on the two-dimensional coordinates of each of the characteristic points C1 to C5 recognized by the image processing means 2, the positional parameter calculation means 3 calculates positional parameters including four parameters that are three-dimensional coordinates (x, y, z) and a pan angle (direction angle) K of the camera 1 with reference to the mark M.
Here, a positional parameter calculation method by the positional parameter calculation means 3 will be described.
First, a point on the ground dropped from a rear axle center O1 of the vehicle 7 vertically with respect to the road surface is set as an origin, a road surface coordinate system in which an x axis and a y axis are set in a horizontal direction and a z axis is set in a vertical direction is assumed, and also, an image coordinate system in which an X axis and a Y axis are set on the image taken by the camera 1 is assumed.
Coordinate values Xm and Ym (m=1 to 5) of the characteristic points C1 to C5 of the mark M in the image coordinate system are expressed by the following expressions from the four positional parameters xm, ym, zm, and K described above using functions F and G.
Xm=F(xm,ym,zm,K)+DXm
Ym=G(xm,ym,zm,K)+DYm
Here, DXm and DYm are deviations between the X coordinates and the Y coordinates of the characteristic points C1 to C5 calculated using the functions F and G, and the coordinate values Xm and Ym of the characteristic points C1 to C5 recognized by the image processing means 2.
In other words, through expression of each of the X coordinates and the Y coordinates of the five characteristic points C1 to C5, ten relational expressions are created in total with respect to the four positional parameters (xm, ym, zm, K).
Therefore, the positional parameters (xm, ym, zm, K) which minimize the following sum of squares of the deviations DXm and DYm are obtained.
S=Σ(DXm2+DYm2)
In other words, an optimization problem that minimizes S is solved. It is possible to use a known optimization method such as a simplex method, a steepest descent method, a Newton method, or a quasi-Newton method.
It should be noted here that the positional parameters are determined by creating relational expressions whose number is greater than the number “four” of the positional parameters (xm, ym, zm, K) to be calculated, so it becomes possible to obtain the positional parameters (xm, ym, zm, K) with accuracy.
In this thirteenth embodiment, ten relational expressions are created for the four positional parameters (xm, ym, zm, K) from the five characteristic points C1 to C5, but it is sufficient that the number of the relational expressions is equal to or greater than the number of the positional parameters (xm, ym, zm, K) to be calculated and when four relational expressions are created from at least two characteristic points, it is possible to calculate the four positional parameters (xm, ym, zm, K).
Also, the parameter zm concerning an attachment height of the camera 1 may be set to a known constant and the remaining three positional parameters that are xm, ym and the pan angle (direction angle) K may be calculated.
Next, in Step S14, travel of the vehicle 7 is started by setting a steering angle of a steering wheel constant and, in Step S15, it is judged whether the vehicle 7 has traveled a predetermined distance from the location A3. When doing so, as to the “predetermined distance”, it is required that a location A4 the vehicle 7 has moved from the location A3 by the predetermined distance is a location at which the mark M enters into the field of view of the camera 1 of the vehicle 7. As to this “predetermined distance”, measurement may be made using a speed pulse signal from the speed sensor 14 or the like, or the driver may travel the vehicle by an appropriate amount at a rough estimate or by intuition. Then, when the vehicle does not yet travel by the predetermined distance, after a steering angle signal from the steering angle sensor 12 is captured in Step S16, a yaw rate signal from the yaw rate sensor 13 is captured in Step S17, and a speed pulse signal from the speed sensor 14 is captured in Step S18, the processing returns to Step S15 and it is judged whether the vehicle has traveled by the predetermined distance. In this manner, during the travel of the vehicle 7 by the predetermined distance, the steering angle signal, the yaw rate signal and the speed pulse signal are repeatedly captured.
When it is judged in Step S15 that the vehicle has traveled by the predetermined distance, the processing proceeds to Step S19 in which the travel of the vehicle 7 is ended and the vehicle 7 is stopped at the location A4. In this state, in Step S20, an image of the mark M is taken by the camera 1 again.
Then, in Step S21, the image processing means 2 extracts the five characteristic points C1 to C5 of the mark M from the image of the mark M taken by the camera 1 and also recognizes and obtains each of two-dimensional coordinates of those characteristic points C1 to C5 on the image. In subsequent Step S22, the positional parameter calculation means 3 calculates the positional parameters including the four parameters that are the three-dimensional coordinates (x, y, z) and the pan angle (direction angle) K of the camera 1 with reference to the mark M on the road surface based on the two-dimensional coordinates of each of the characteristic points C1 to C5 recognized by the image processing means 2.
After the positional parameters at two locations that are the location A3 and the location A4 are calculated in this manner, the processing proceeds to Step S23 in which the vehicle travel parameter calculation means 15 calculates a turn radius R, a turn angle θ and a moving distance AR of the vehicle 7 corresponding to the movement from the location A3 to the location A4 based on the positional parameters at the two locations calculated in Steps S13 and S22.
Here, a method of calculating the turn radius R, the turn angle θ and the moving distance AR will be described using
The positional parameters calculated by the positional parameter calculation means 3 include the four parameters that are the three-dimensional coordinates (x, y, z) and the pan angle (direction angle) K of the camera 1 with reference to the mark M on the road surface, so it becomes possible to grasp positions and directions of the vehicle 7 at both of the locations A3 and A4. Therefore, at the location A3, a straight line SL1 which passes through the rear axle center O1 of the vehicle 7 and is vertical to a center line CL1 of the vehicle 7 is calculated. Similarly, at the location A4, a straight line SL2 which passes through the rear axle center O2 of the vehicle 7 and is vertical to a center line CL2 of the vehicle 7 is calculated. Then, when an intersection between those straight lines SL1 and SL2 is obtained, this becomes a turn center CP of the vehicle 7, and when an intersecting angle between the straight lines SL1 and SL2 is obtained, this becomes the turn angle θ of the vehicle 7. Also, when a distance from the turn center CP to the rear axle center O1 or O2 of the vehicle 7 at the location A3 or A4 is calculated, this becomes the turn radius R. A turn circular arc Q drawn by the movement of the vehicle 7 due to the coordinate of the turn center CP and the turn radius R is calculated and a circular arc length of this circular arc Q with respect to the turn angle θ becomes the moving distance AR of the vehicle 7.
In subsequent Step S24, the vehicle travel parameter calculation means 15 calculates the turn radius R, the turn angle θ and the moving distance AR of the vehicle 7 corresponding to the movement from the location A3 to the location A4 based on the steering angle signal, the yaw rate signal and the speed pulse signal captured in Steps S16 to 18.
The turn radius R of the vehicle 7 with respect to the steering angle is set in advance in the vehicle 7 in a map form or using a relational expression, and the vehicle travel parameter calculation means 15 calculates the turn radius R of the vehicle 7 using the map or relational expression described above based on the steering angle signal from the steering angle sensor 12.
Also, through an integration process of the yaw rate signal from the yaw rate sensor 13 and a multiplication by a gain of the yaw rate sensor 13 set in advance, a yaw angle of the vehicle 7 is detected. Therefore, by obtaining a difference between the yaw angles at both of the locations A3 and A4, the turn angle θ of the vehicle 7 from the location A3 to the location A4 is calculated.
In addition, the moving distance AR of the vehicle 7 is calculated, by multiplying the number of pulses of the speed pulse signal obtained by the speed sensor 14 from the location A3 to the location A4 by a moving distance per speed pulse set in advance.
Finally, in Step S25, the vehicle travel parameter calculation means 15 calculates the travel parameters of the vehicle 7 by comparing the turn radius R, the turn angle θ and the moving distance AR calculated from the positional parameters of the camera 1 in Step S23 and the turn radius R, the turn angle θ and the moving distance AR calculated from the detection signals of the various sensors in Step S24 with each other.
In other words, the map or relational expression of the turn radius R with respect to the steering angle is calculated or the map or relational expression of the turn radius R with respect to the steering angle set in advance is corrected so that a value of the turn radius R obtained in Step S24 becomes a value of the turn radius R obtained in Step S23.
Also, the gain of the yaw rate sensor 13 is calculated or the gain of the yaw rate sensor 13 set in advance is corrected so that a value of the turn angle θ obtained in Step S24 becomes a value of the turn angle θ obtained in Step S23.
Further, the moving distance per speed pulse is calculated or the moving distance per speed pulse set in advance is corrected so that a value of the moving distance AR obtained in Step S24 becomes a value of the moving distance AR obtained in Step S23.
It should be noted here that it is possible to construct the image processing means 2, the positional parameter calculation means 3 and the vehicle travel parameter calculation means 15 from a computer, and by setting a vehicle travel parameter calculation program of the operations in Steps S11 to S25 of
Also, it is possible to collectively form the vehicle travel parameter calculation apparatus part P2 constructed by the input portion K, the image processing means 2, the positional parameter calculation means 3 and the vehicle travel parameter calculation means 15 in a form of a substrate module, a chip, or the like. A vehicle travel parameter calculation apparatus is realized merely by connecting the camera 1 mounted on the vehicle to the input portion K of this vehicle travel parameter calculation apparatus part P2 and connecting the steering angle sensor 12, the yaw rate sensor 13 and the speed sensor 14 to the vehicle travel parameter calculation means 15.
In the thirteenth embodiment described above, the turn radius R, the turn angle θ and the moving distance AR are calculated based on the positions and the directions of the vehicle 7 at the two locations A3 and A4, but when the position of the vehicle 7 at each of three locations is known, it is possible to identity a circular arc orbit of turn, so it is also possible to calculate those turn radius R, turn angle θ and moving distance AR from the position of the vehicle 7 at the three locations.
Also, in the thirteenth embodiment described above, the turn radius R with respect to a steering angle, the gain of the yaw rate sensor 13 and the moving distance per speed pulse are each calculated as the vehicle travel parameter, but a construction in which only any one or two of those are calculated is also possible.
An image of the mark M is taken by the camera 1 of the vehicle 7 in a stop state at each of the two locations A3 and A4, but it is sufficient that the vehicle 7 moves between the location A3 and the location A4, and an image of the mark M may be taken at each of two locations during the travel of the vehicle 7.
Also, the figure shown in
Further, when a mark M having three or more characteristic points is used, it is possible to create six or more relational expressions by expressing each of an X coordinate and a Y coordinate in an image coordinate system of each characteristic point, so it becomes possible to calculate positional parameters of the camera 1 including six parameters that are the three-dimensional coordinates (x, y, z), a tilt angle (dip angle), the pan angle (direction angle) and a swing angle (rotation angle). As a result, it becomes possible to calculate the turn radius R, the turn angle θ and the moving distance AR of the vehicle 7 with accuracy even when there is a difference of altitude of the road surface or the like, thereby improving calculation accuracy of the travel parameters of the vehicle 7.
In the thirteenth embodiment described above, such a construction with only one camera 1 has been described. However, it is also possible to mount two cameras, at least a part of whose fields of view overlap each other, on the vehicle 7 and simultaneously take images of the mark M in the overlapping fields of view with both of the cameras. In this case, it is possible to create four relational expressions from one characteristic point, so it becomes possible to calculate positional parameters of the camera 1 including four parameters that are the three-dimensional coordinates (x, y, z) and the pan angle (direction angle) when the mark M has one characteristic point, and it becomes possible to calculate positional parameters of the camera 1 including six parameters that are the three-dimensional coordinates (x, y, z), the tilt angle (dip angle), the pan angle (direction angle) and the swing angle (rotation angle) when there are two characteristic points. In addition, a construction with three or more cameras is also possible.
It is possible to calculate the travel parameters of the vehicle 7 by taking an image of the mark M at each of more locations including the two locations A3 and A4, and also repeatedly capturing the detection signals from the various sensors between the locations. In this case, it is sufficient that, at the many locations, the travel parameters such as the turn radius R with respect to a steering angle, the gain of the yaw rate sensor 13 and the moving distance per speed pulse are calculated or corrected so that the values of the turn radius R, the turn angle θ and the moving distance AR obtained from the detection signals of the various sensors become the most rational values with respect to the values of the turn radius R, the turn angle θ and the moving distance AR calculated from the positional parameters of the camera 1.
In addition, it is also possible to calculate the travel parameters of the vehicle 7 by continuously taking images of the mark M and also capturing the detection signals from the various sensors during the travel of the vehicle 7. In this case, the travel parameters may be calculated or corrected based on a vehicle behavior corresponding to the changing steering angle by moving the vehicle while changing the steering angle.
In the thirteenth embodiment described above, the mark M is arranged on the road surface and is set as the fixed target outside the vehicle, but it is also possible to, as shown in
In the thirteenth and fourteenth embodiments described above, the camera 1 is embedded in the door mirror 8 positioned in a side portion of the vehicle 7, but the present invention is not limited to this. For instance, the camera 1 may be installed in a rear portion of the vehicle 7 to take an image behind the vehicle 7.
When the travel parameters are obtained by turning the vehicle 7, it is preferable that the travel parameters are calculated or corrected independently between left turn and right turn. In addition, there is also a case where the moving distances per speed pulse are different between at the time of turn and at the time of straight advance, so it is preferable that not only a value at the time of turn travel but also a value at the time when the vehicle 7 is traveled straight ahead is calculated.
A construction of a parking assistance apparatus according to a fifteenth embodiment is shown in
In the thirteenth and fourteenth embodiments, the vehicle 7 is traveled in accordance with a special sequence for calculating the vehicle travel parameters and an image of the mark M or the lattice figure N is taken, but in this fifteenth embodiment, the vehicle travel parameters are calculated by the vehicle travel parameter calculation means 15 during parking of the vehicle 7 into the parking space based on the guide information provided from the guide apparatus 6.
First, a parking assist is performed in the same manner as in the operation in the first embodiment shown in
The calculated positional parameters are sent to the vehicle travel parameter calculation means 15 and are also sent to the relative position identification means 4, and the relative positional relation between the vehicle 7 and the parking space S is identified by the relative position identification means 4. In addition, the parking locus for leading the vehicle 7 to the parking space S is calculated by the parking locus calculation means 5 based on this relative positional relation, and the guide information is created by the guide information creation means 10 of the guide apparatus 6 and is outputted from the guide information output means 11 to the driver.
When travel of the vehicle 7 is started in accordance with the guide information, the vehicle travel parameter calculation means 15 repeatedly captures the steering angle signal from the steering angle sensor 12, the yaw rate signal from the yaw rate sensor 13 and the speed pulse signal from the speed sensor 14, and measures a moving distance of the vehicle 7 based on those signals, and an image of the mark M is taken again by the camera 1 at a location at which the vehicle has traveled by a predetermined distance. Then, two-dimensional coordinates on the image of the characteristic points C1 to C5 of the mark M are recognized by the image processing means 2, and the positional parameters of the camera 1 are calculated by the positional parameter calculation means 3 and are sent to the vehicle travel parameter calculation means 15.
After the positional parameters of the camera 1 at two locations are sent to the vehicle travel parameter calculation means 15 in this manner, the vehicle travel parameter calculation means 15 calculates the turn radius R, the turn angle θ and the moving distance AR of the vehicle 7 corresponding to a movement between the two locations based on those positional parameters.
Next, the vehicle travel parameter calculation means 15 calculates the turn radius R, the turn angle θ and the moving distance AR of the vehicle 7 corresponding to the movement between the two locations based on the repeatedly captured steering angle signal from the steering angle sensor 12, yaw rate signal from the yaw rate sensor 13 and speed pulse signal from the speed sensor 14.
In addition, the travel parameters of the vehicle 7, such as the turn radius R with respect to a steering angle, the gain of the yaw rate sensor 13 and the moving distance per speed pulse, are calculated through comparison between the turn radius R, the turn angle θ and the moving distance AR calculated from the positional parameters of the camera 1 and the turn radius R, the turn angle θ and the moving distance AR calculated from the detection signals of the various sensors.
The calculated travel parameters are sent from the vehicle travel parameter calculation means 15 to the guide information creation means 10 of the guide apparatus 6 and are updated.
As described above, it is possible to carry out the calculation of the travel parameters of the vehicle 7 in a parking sequence based on the guide information and it is also possible for the guide information creation means 10 to create the guide information using the calculated travel parameters, so it becomes possible to perform a highly accurate parking guide even when parking into the parking space S is performed for the first time.
It is possible to construct the image processing means 2, the positional parameter calculation means 3, the relative position identification means 4, the parking locus calculation means 5 and the vehicle travel parameter calculation means 15 from a computer, and by setting a parking assist program of the operations described above to the computer from a recording medium or the like in which the program is recorded, it becomes possible to cause the computer to execute each step.
Also, a parking assistance apparatus part P3 is constructed by the input portion K, the image processing means 2, the positional parameter calculation means 3, the relative position identification means 4, the parking locus calculation means 5 and the vehicle travel parameter calculation means 15 and it is possible to collectively form this parking assistance apparatus part P3 in a form of a substrate module, a chip, or the like.
A construction of a parking assistance apparatus according to a sixteenth embodiment is shown in
An operation of the sixteenth embodiment is shown in a flowchart in
It should be noted here that also in the second to twelfth embodiments, it is possible to perform the automatic steering by similarly applying the sixteenth embodiment.
A construction of a parking assistance apparatus according to a seventeenth embodiment is shown in
The vehicle travel parameters such as the turn radius of the vehicle 7 with respect to a steering angle, the gain of the yaw rate sensor 13 and the moving distance per speed pulse are set in advance in the automatic steering apparatus 16. Based on the detection signals from the steering angle sensor 12, the yaw rate sensor 13 and the speed sensor 14 and the parking locus calculated by the parking locus calculation means 5, the automatic steering apparatus 16 creates a steering signal for automatically steering a steering wheel so that the vehicle 7 is capable of traveling along the parking locus.
Then, during a movement of the vehicle 7 into the parking space S through a brake operation and an acceleration operation in which steering is performed by the automatic steering apparatus 16, the vehicle travel parameters are calculated by the vehicle travel parameter calculation means 15, are sent from the vehicle travel parameter calculation means 15 to the automatic steering apparatus 16, and are updated. As a result, it becomes possible to perform highly accurate parking.
A construction of a parking assistance apparatus according to an eighteenth embodiment is shown in
An operation of the eighteenth embodiment is shown in a flowchart in
It should be noted here that also in the second to twelfth embodiments, it is possible to perform the automatic parking by similarly applying the eighteenth embodiment.
A construction of a parking assistance apparatus according to a nineteenth embodiment is shown in
The vehicle travel parameters such as the turn radius of the vehicle 7 with respect to a steering angle, the gain of the yaw rate sensor 13 and the moving distance per speed pulse are set in advance in the automatic travel apparatus 17. Based on the detection signals from the steering angle sensor 12, the yaw rate sensor 13, and the speed sensor 14 and the parking locus calculated by the parking locus calculation means 5, the automatic travel apparatus 17 creates a travel signal for causing the vehicle 7 to automatically travel along the parking locus.
Then, during automatic travel of the vehicle 7 into the parking space S by the automatic travel apparatus 17, the vehicle travel parameters are calculated by the vehicle travel parameter calculation means 15, are sent from the vehicle travel parameter calculation means 15 to the automatic travel apparatus 17, and are updated. As a result, it becomes possible to perform highly accurate automatic parking.
In each embodiment described above, when an obstacle sensor such as an ultrasonic sensor is mounted on the vehicle 7 and a warning is issued or an obstacle avoidance operation is performed in the case where the existence of a peripheral obstacle is recognized, a safer parking assist is provided.
It is also possible to use an object, such as a sprag or a pattern of a wall surface of a garage, which originally exists on the periphery of the parking space as the fixed target instead of installing the mark at a predetermined place having a predetermined positional relation with respect to the parking space. However, it is preferable that the existence of the object is easy to perceive and characteristic points internally included in the object be easy to recognize.
When a sensor that detects a vehicle height is provided to the vehicle 7, it becomes possible to compensate for a change of an installation height of the camera due to an increase/decrease of passengers, a fuel or a load, a secular change of a suspension, or the like.
In the fourth embodiment, it is also possible to provide a moving amount sensor that detects a moving distance and a moving direction for the vehicle 7 and correct, when there is an error between a predicted vehicle position and a recognized vehicle position by the mark M, the parameters (such as the turn radius with respect to a steering angle, the moving distance per speed pulse and the gain of the yaw rate sensor) of the vehicle 7 so that the error is eliminated. Also when there is a difference between left-side parking and right-side parking, it is preferable that a correction be made by distinguishing between the left and the right. After the correction, there hardly occurs an error of a calculated orbit, so an orbit, along which the vehicle actually travels, becomes a smooth orbit including no meander or the like, which makes it possible to perform safe and highly accurate parking. It is not required to make this correction at each time of a parking operation and it is sufficient that the correction is carried out at appropriate cycles. Also, the cycles may be determined in accordance with a distance between the mark and the vehicle 7. For instance, when the distance is long, the cycles for carrying out the correction are elongated, whereby a load of computation is reduced.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2006-355498 | Dec 2006 | JP | national |
| 2007-005984 | Jan 2007 | JP | national |
| 2007-119359 | Apr 2007 | JP | national |
| 2007-260800 | Oct 2007 | JP | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/JP2007/072358 | 11/19/2007 | WO | 00 | 10/12/2009 |
