JP2009250718A - Vehicle position detecting apparatus and vehicle position detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a vehicle position detecting apparatus, capable of detecting and correcting the position of own vehicle, with a very high degree of accuracy by taking the distance to a target into consideration. <P>SOLUTION: A navigation system for the vehicle stores beforehand the information, including the position, shape, color and letters of "kilo-posts" placed on a road in a storage part 15c; performs image recognition of the kilo-post as a target by an image recognition part 15a from the image taken by a camera 11 for taking images around own vehicle, based on the information stored in the storage part 15c; and corrects the position of own vehicle, by moving the map position of own vehicle displayed on a display device 14 to the position of kilo-post, obtained by the image recognition by the image recognition part 15a, and stored in the storage part 15c. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a vehicle position detection device and a vehicle position detection method for detecting the position of a vehicle.

  The vehicle navigation system displays a map on a display mounted in the vehicle and displays the position of the host vehicle on the map, thereby allowing the driver to know the location where the host vehicle is traveling. Therefore, the vehicle navigation system is required to detect the own vehicle position with high accuracy in order to notify the driver of the exact own vehicle position. If the detection accuracy of the host vehicle position is poor, the actual driving location differs from the host vehicle position displayed on the display, and accurate information cannot be presented to the driver.

Therefore, as a technique for improving the vehicle position detection accuracy, for example, a technique described in Patent Document 1 has been proposed. Specifically, in Patent Document 1, the accurate position of a target such as a sign or a signboard is stored in a database in advance, and when the target is recognized by a camera mounted on a vehicle, the target is stored in the database. It is disclosed that the position of the host vehicle is corrected using certain position information.
JP 2000-97714 A

  However, in the conventional technique described in Patent Document 1 described above, since the target vehicle is recognized and the position of the host vehicle is corrected, the distance to the target is not considered, and the host vehicle and the target are not considered. There is a problem that an error occurs by the distance of.

  Therefore, the present invention has been proposed in view of the above-described circumstances, and a vehicle position detection device and a vehicle position detection capable of correcting the position of the host vehicle with extremely high accuracy in consideration of the distance to the target object. It aims to provide a method.

  In the present invention, when the position of the vehicle is detected, image recognition based on the characteristics of the target is performed on an image captured by an imaging unit that captures an image around the host vehicle, and the image of the target that has been recognized is imaged. Based on the position in the video, the host vehicle position on the map displayed on the display means is moved to correct the host vehicle position.

  According to the present invention, the position of the host vehicle is corrected by recognizing the target actually installed on the road, so that the change in the position of the target in the image according to the distance to the target is taken into consideration. Thus, the position of the host vehicle can be detected and corrected with extremely high accuracy.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  The present invention is applied to a vehicle navigation system configured as shown in FIG. 1, for example.

[Configuration of vehicle navigation system]
As shown in FIG. 1, a vehicle navigation system shown as an embodiment of the present invention includes a camera 11 that captures an image around a host vehicle, a GPS receiver 12 that receives a signal from a GPS (Global Positioning System) satellite, A vehicle speed sensor 13 for detecting the speed of the host vehicle, a display device 14 for displaying route guidance information, and a navigation body 15 for performing route guidance for the host vehicle are provided.

  The camera 11 is attached, for example, in front of the host vehicle and captures an image around the host vehicle. In addition, the attachment position of the camera 11 can be set as the arbitrary place of the own vehicle, if it is a location which can image the front of the own vehicle. The video imaged by the camera 11 is supplied to the navigation main body 15.

  The GPS receiver 12 receives radio waves from GPS satellites and supplies position information based on the obtained GPS signals to the navigation main body 15.

  The vehicle speed sensor 13 detects the speed of the host vehicle and supplies the detected speed information to the navigation body 15.

  The display device 14 is composed of, for example, a liquid crystal display or the like mounted in the vehicle interior, and displays route guidance information including a map output from the navigation main body 15 and the vehicle position.

  The navigation body 15 includes an image recognition unit 15a, a host vehicle position detection unit 15b, a storage unit 15c, and a video output unit 15d. In FIG. 1, the navigation main body 15 is configured by computer hardware including a CPU, a ROM, a RAM, and the like. However, in FIG. 1, for convenience, the description is given separately for each functional block.

  For example, as shown in FIG. 2, the image recognition unit 15a performs pattern recognition of a kilopost as a target based on feature information such as the shape, color, and characters of the kilopost stored in the storage unit 15c. Part 15a ′. The image recognition unit 15a including the pattern recognition unit 15a 'performs pattern recognition processing at a predetermined cycle such as a cycle of 100 milliseconds. Then, the image recognition unit 15a recognizes information indicating whether or not the recognized number of kiloposts is a target number and information indicating the position of the kilopost on the imaging screen for each period of the recognition processing as a recognition result. It supplies to the own vehicle position detection part 15b.

  For example, as shown in FIG. 3A, when the kilometer post 111 is on the left shoulder of the road 110 and the processing period of the pattern recognition processing by the image recognition unit 15a is 100 milliseconds, the image is taken by the camera 11. Since the range to be processed is constant, the locus of the kilopost 111 recognized by the image recognition unit 15a every 100 milliseconds is as shown in FIG. Then, as shown in FIG. 3B, the image recognition unit 15a obtains the image heights y0, y1, y2, y3, y4, y5, and y6 of the kilopost 111 within the range imaged by the camera 11, respectively. The result is supplied to the own vehicle position detector 15b.

  The own vehicle position detection unit 15b detects the position of the own vehicle based on the recognition result of the image recognition unit 15a, the position information supplied from the GPS receiver 12, and the speed information supplied from the vehicle speed sensor 13. to correct. The own vehicle position detection unit 15b supplies the detected position information of the own vehicle to the video output unit 15d.

  The storage unit 15c is made of a storage medium such as a hard disk, a DVD, or a CD, for example. In addition to map data used for route guidance of the host vehicle, the location information of the kiloposts installed on each expressway, Predetermined information 15c ′ including feature information of kiloposts including shape, color, and characters is stored.

  Here, the vehicle navigation system may store the position information and the feature information about all the kilometer posts existing on the expressway in the storage unit 15c, but in order to reduce the data amount, Only the position information and the feature information may be stored in the storage unit 15c. Specifically, as shown in FIG. 4, for example, as shown in FIG. 4, the storage unit 15c is divided into an up line (a) and a down line (b) of a certain highway, such as 10 kilo points, 20 kilo points, and 30 kilo points. As described above, only the position information and feature information about the kilometer post installed every 10 kilometers may be stored.

  When the vehicle navigation system detects that the host vehicle is approaching within a predetermined distance from the nearest kilo point by storing only the position information and characteristic information about the specific kilo post in the storage unit 15c in this way. As long as it is possible, the image recognition processing by the image recognition unit 15a can be started. Thus, it is possible to prevent an increase in processing load due to always performing image processing. The map data stored in the storage unit 15c is read by the video output unit 15d, and the feature information of the kilopost is read by the image recognition unit 15a.

  The video output unit 15d creates a video for performing route guidance based on the map data read from the storage unit 15c and the position information of the host vehicle supplied from the host vehicle position detection unit 15b. Output to the display device 14.

  Such a navigation main body 15 creates a video as route guidance information in order to display a map or a vehicle position on the map as a vehicle mark on the display device 14, and outputs the video to the display device 14. The route guidance of the own vehicle is performed.

  In such a vehicle navigation system, the kilometer post feature information including the position information, shape, color, and characters of the kilometer post installed on the expressway is stored in the storage unit 15c in advance, and the vehicle is running. When the position of the host vehicle is detected, the corresponding kilometer post is imaged by the camera 11. And the image recognition part 15a performs image recognition with respect to the picked-up image of the camera 11, and correct | amends the own vehicle position with high precision.

[Operation of vehicle navigation system]
In the existing vehicle navigation system, when the own vehicle position is specified by the autonomous navigation, various errors occur, so the actual driving position and the own vehicle position on the map are different. There are many cases. In addition, in existing vehicle navigation systems, on general roads with intersections and sharp curves, so-called map matching technology is used to correct the position of the vehicle, but on expressways with no intersections or sharp curves, The host vehicle position cannot be corrected by the map matching technique. For example, as shown in FIG. 5, the error accumulates as the host vehicle 120 travels. The position will be very different. As a result, the vehicle position 101 on the display screen 100 is far from the correct vehicle position 101 ′.

  There is also a technique for correcting the position of the host vehicle based on the position information. In the first place, the position information received from the GPS satellite includes an error. In general, the error of the position information is about 10 m when the vehicle is stopped, and it is said that the accuracy further decreases during traveling. Therefore, the existing vehicle navigation system can correctly identify the position of the host vehicle when there is no error in the position information acquired from the GPS receiver 12, but the host vehicle using the existing GPS receiver 12 can be specified. Even if the position is corrected, there is an error of 10 m or more, and the position of the host vehicle 120 cannot be corrected correctly.

  On the other hand, the vehicle navigation system to which the present invention is applied starts the image recognition process according to a series of procedures as shown in FIG.

  First, as shown in FIG. 6, the navigation main body 15 detects the host vehicle position by the host vehicle position detection unit 15b in step S1. At this time, the host vehicle position detection unit 15b may specify the latitude and longitude of the host vehicle position using the GPS signal from the GPS receiver 12.

  Subsequently, in step S2, the own vehicle position detection unit 15b refers to the map data stored in the storage unit 15c and determines whether the currently traveling road is a highway or a general road. . If the currently running road is a general road, the process of step S2 is repeated. On the other hand, if the currently running road is a highway, the storage unit in step S3 Based on the kilo-post information stored in 15c, the nearest kilo-post is determined from the current position.

  For example, when the information 15c ′ of the kilometer post 121 in the situation as shown in FIG. 7A is stored in the storage unit 15c, the entrance to the expressway is near “16 kilometers” of the kilometer post 121. In this case, the host vehicle position detection unit 15b determines that the nearest kilometer post 121 is “20 kilometers”.

  Then, in step S4, the own vehicle position detection unit 15b determines the nearest position from the position of the own vehicle 120 based on the position information of the nearest kilopost 121 stored in the storage unit 15c and the detected position of the own vehicle 120. The distance to the kilometer post 121 is calculated. Then, when the host vehicle 120 travels and the situation shown in FIG. 7B is reached, the calculated distance becomes equal to or less than the predetermined distance L. In step S5, the surrounding image is captured by the camera 11, and the image recognition unit 15a Start the image recognition process.

  By starting image recognition processing according to such a series of procedures, the vehicle navigation system need not always perform image recognition, and can prevent an increase in processing load.

  Next, processing from the start of imaging by the camera 11 to the correction of the vehicle position will be described with reference to FIGS.

  First, when it is detected by the navigation body 15 that the distance to the target kilometer post 121 has reached the predetermined distance L, the camera 11 starts capturing surrounding images in step S11 as shown in FIG. .

  Subsequently, in step S12, the image recognition unit 15a in the navigation main body 15 receives the image captured by the camera 11 and the feature information including the shape, color, and characters of the kilopost 121 stored in the storage unit 15c in advance. Compare and start pattern recognition of kiloposts as targets. As this characteristic information, the character, color, and shape of the kilopost 121 closest to the position of the host vehicle measured by the GPS receiver 12 is selected and used for pattern recognition.

  And the navigation main body 15 advances a process to step S14, when the kilometer post 121 as a target object can be recognized by the image recognition part 15a in step S13.

  In step S14, the position of the host vehicle mark on the map displayed on the display device 14 by the host vehicle position detecting unit 15b and the video output unit 15d is used as a kilometer post as a target stored in advance in the storage unit 15c. An image moved to the position 121 is created, and the vehicle position is corrected. Specifically, for example, as shown in FIG. 9A, the navigation main body 15 shows that the kilometer post 121 recognized as the target indicates “20 km” as shown in FIG. 9B. Then, the position of the host vehicle mark 101 ′ on the map is moved to the position of the kilometer post 121, and the position of the host vehicle mark 101 displayed with an error is corrected.

  In the vehicle navigation system, the position of the host vehicle can be corrected with high accuracy according to such a series of procedures.

  In the navigation main body 15, when the kilometer post as the target passes directly beside the host vehicle, that is, when the host vehicle reaches the same position as the kilometer post as the target, the host vehicle position is corrected. desirable.

  However, in the navigation main body 15, when the angle of view of the camera 11 is narrow, for example, in the state shown in FIG. 10A, the kilopost 121 is within the imaging range 120 ′ and the road as shown in FIG. Although the captured image 100 including the 110 kilometer posts 111 can be obtained, when the host vehicle 120 approaches the kilometer post 121 as shown in FIG. 10C, the kilometer post 121 is captured within the imaging range 120 ′ as shown in FIG. It will be outside. As a result, as shown in FIG. 10D, the captured image 100 does not include the kilopost 111, so that image recognition cannot be performed and the vehicle position cannot be corrected at an optimal timing. . Further, in the navigation main body 15, even if the field of view of the camera 11 is wide and the kilometer post located just beside the host vehicle 120 can be imaged, the speed of movement of the kilopost within the imaging screen of the camera 11 as the kilopost approaches the host vehicle. It is also conceivable that the recognition rate of image recognition decreases.

  Therefore, in the navigation main body 15, it is estimated that the kilometer post as the target arrives directly beside the host vehicle, and the host vehicle position is corrected.

  As described above, the image recognition unit 15a detects the trajectory for each cycle of the image processing from when the kilometer post as the target is first recognized until the kilopost is outside the range of the captured image of the camera 11. Calculate the height of the kilometer post within the imaging range. Therefore, the kilopost height y0 at the time t0 when the kilopost was first recognized, the kilopost height y1 at the time t1 after one cycle, the kilopost height y2 at the time t2 after the first cycle, and so on. The height within the imaging range is known.

  Here, when the kilometer post passes directly beside the host vehicle, it can be considered that the height Y of the kilometer post captured by the camera 11 is substantially equal to zero.

  Therefore, in the navigation main body 15, for example, when the kilopost 111 moves in the captured image 100 as shown in FIG. 11A, the height of the kilopost is calculated by the image recognition unit 15a as shown in FIG. 11B. The time T and its height Y are acquired. Then, an approximate expression is obtained using the time T and the height Y, and the time T at which the height Y = 0 is calculated, so that the time T is determined as a timing for correcting the vehicle position. Thereby, the navigation main body 15 estimates the timing at which the positions of the host vehicle and the target are equivalent based on the trajectory of the target detected every predetermined period. At the estimated timing, the navigation main body 15 can correct the position of the host vehicle.

  In the vehicle navigation system, for example, there is no need to correct the position of the host vehicle by obtaining the distance to the target using a stereo camera composed of two cameras. A highly accurate correction of the vehicle position can be realized.

  In addition, although the case where the kilometer post installed on the road was set as the target was described above, in the vehicle navigation system, the guide sign installed at the exit of the expressway, the sign of the speed limit, the expressway Arbitrary structures such as bridges erected above can be used as the target.

[Effect of the embodiment]
As described above in detail, in the vehicle navigation system shown as the embodiment of the present invention, the storage unit 15c stores in advance the feature information including the position information, shape, color, and characters of the kilometer post installed on the road. The image is recognized by the image recognition unit 15a from the image captured by the camera 11, and the vehicle on the map displayed on the display device 14 based on the position in the image of the kilopost. The position of the host vehicle is corrected by moving the position to the position of the kilometer post recognized by the host vehicle position detection unit 15b and the video output unit 15d and stored in the storage unit 15c.

  Thereby, according to this vehicle navigation system, the position of the host vehicle can be detected and corrected with extremely high accuracy in consideration of the change in the position of the kilometer post in the image according to the distance to the kilometer post as the target. Can do.

  Further, in this vehicle navigation system, the navigation main body 15 corrects the position of the host vehicle at the timing when the kilopost passes right next to the host vehicle based on the position in the captured image of the kilopost. In addition, the navigation main body 15 estimates the timing at which the position of the host vehicle and the kilometer post is equivalent based on the locus of the kilometer post as the target detected by the image recognizing unit 15a every predetermined period. The own vehicle position is corrected. As a result, the distance to the target can be obtained without using an expensive stereo camera, and even with a low-cost configuration, the vehicle position correction with higher accuracy can be realized.

  Furthermore, in this vehicle navigation system, the height T of the kilopost within the imaging range of the camera 11 is calculated at predetermined intervals by the image recognition unit 15a, and the time T when the height of the kilopost is calculated by the image recognition unit 15a and the time T By determining the timing for correcting the position of the own vehicle based on the approximate expression obtained using the height Y, it is possible to realize more accurate correction of the own vehicle position.

  Furthermore, in this vehicle navigation system, the distance from the vehicle position to the kilometer post is calculated based on the position information of the kilometer post as the target stored in the storage unit 15c and the detected vehicle position. When the calculated distance is equal to or less than the predetermined distance L, the image recognition processing by the image recognition unit 15a is started, so that it is not always necessary to perform image processing, and an increase in processing load can be prevented.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and even if it is a form other than this embodiment, as long as it does not depart from the technical idea according to the present invention, the design and the like Of course, various modifications are possible.

It is a block diagram shown about the composition of the navigation system for vehicles shown as an embodiment of the present invention. It is a block diagram for demonstrating the processing content of the image recognition part in the navigation system for vehicles shown as embodiment of this invention. It is a figure for demonstrating a mode that the height of the kilopost within the imaging range of a camera is calculated | required by the image recognition part in the vehicle navigation system shown as embodiment of this invention. It is a figure for demonstrating the content of the kilometer post | mailbox information memorize | stored in the memory | storage part in the navigation system for vehicles shown as embodiment of this invention. It is a figure for demonstrating the error of the place which is actually drive | working, and the own vehicle position. It is a flowchart which shows a series of procedures until it starts an image recognition process in the vehicle navigation system shown as embodiment of this invention. It is a figure for demonstrating a mode that the nearest kilometer post is determined in the vehicle navigation system shown as embodiment of this invention. 5 is a flowchart showing a series of procedures from the start of imaging by a camera to the correction of the vehicle position in the vehicle navigation system shown as the embodiment of the present invention. It is a figure for demonstrating a mode that the position of the own vehicle mark on a map is moved to the position of the kilometer post recognized as a target, and the own vehicle position is correct | amended in the vehicle navigation system shown as embodiment of this invention. It is a figure for demonstrating a mode that a kilopost is outside an imaging range. It is a figure for demonstrating a mode that an approximate expression is calculated | required using the time calculated by the image recognition part, and the height of a kilopost in the vehicle navigation system shown as embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Camera 12 GPS receiver 13 Vehicle speed sensor 14 Display apparatus 15 Navigation main body 15a Image recognition part 15b Own vehicle position detection part 15c Memory | storage part 15d Image | video output part

Claims (6)

A vehicle position detection device that detects the position of the host vehicle and displays the position on a display means,
Storage means for storing the position of the target installed on the road, feature information representing the feature of the target,
Image recognition means for recognizing the target image from the captured image captured by the imaging means for capturing the image around the host vehicle using the feature information of the target stored in the storage means;
Based on the position in the captured image of the target recognized by the image recognition means, the position of the target recognized by the image recognition means and the position of the target stored in the storage means. A vehicle position detection apparatus comprising: correction means for performing correction for moving the position of the host vehicle displayed on the display means.   The correction unit corrects the position of the host vehicle at a timing when the target passes right next to the host vehicle based on the position in the captured image of the target recognized by the image recognition unit. The vehicle position detection device according to claim 1.   The correction means estimates the timing at which the positions of the host vehicle and the target are equivalent based on a trajectory in the captured image of the target detected at predetermined intervals by the image recognition means, and performs the estimation The vehicle position detection device according to claim 1, wherein the position of the host vehicle is corrected at the determined timing. The image recognition means calculates the height of the target in the captured video for each predetermined period,
The correction means determines a timing for correcting the position of the host vehicle based on a time when the height of the target is calculated by the image recognition means and an approximate expression obtained using the height. The vehicle position detection device according to claim 3.   The correction unit calculates a distance from the vehicle position to the target based on the position of the target stored in the storage unit and the detected vehicle position, and the calculated distance is a predetermined distance. 5. The vehicle position detection device according to claim 1, wherein an image recognition process by the image recognition unit is started when the following occurs. When detecting the position of the vehicle,
A map that is displayed on the display unit based on the position of the target image in the captured image of the target image that has been recognized by the imaging unit that captures the video around the host vehicle. A vehicle position detection method, wherein the position of the host vehicle is corrected by moving the host vehicle position above.

Images