JPH10300494A - Vehicle position detecting device and method and route guide device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To guide a vehicle on a route in detailed fashion by enhancing positioning accuracy using traffic lane information while the vehicle is traveling on a road. SOLUTION: A navigation system 10 and a traveling environment recognition device 11 are mounted in a vehicle. A vehicle position detecting device 12 detects the position of the vehicle on the basis of sensors such as a GPS. Because the vehicle position detected can be erroneous to a great extent, the position is corrected on the basis of road map data stored in a map CD-ROM in advance. Information about the number of traffic lanes on the road is also stored in the map CD-ROM in advance. The road on which the vehicle is traveling is photographed by a video camera 15, and the traffic lanes are recognized by an image recognition device 16. The number of the traffic lanes is determined according to the traffic lanes recognized, and the vehicle position detecting device 12 determines that the vehicle is traveling on the road where the number of traffic lanes is equal to the number determined. Even while the vehicle is running on a road overlapping with an elevated road, the road on which the vehicle is traveling can be determined correctly according to the number of the traffic lanes.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle position detecting device and method mounted on a vehicle such as an automobile running on a road and used for detecting a current position, and route guidance for guiding a route on which the vehicle travels. Apparatus and method.

[0002]

2. Description of the Related Art Conventionally, a vehicle such as an automobile running on a road has a current position of the vehicle on a road map based on road map data stored in advance on a recording medium such as a compact disk (CD). , And a navigation device having a function of guiding a vehicle along a route set in advance. The prior art regarding the navigation device is disclosed in, for example, Japanese Patent Application Laid-Open No. H7-198401.

FIG. 13 shows an example of route guidance performed by a conventional navigation device. Road map display 1
For example, using the entire screen. On a specific road on the road map display 1, a vehicle position 2 indicating the current position of the vehicle is displayed. When guidance is provided along the guidance route 3 set in advance for the vehicle, the guidance point 4 may also be displayed. The guide point 4 is set in advance, for example, at an intersection where the vehicle changes the traveling direction.
When the vehicle changes the traveling direction at the guide point 4, the destination direction display 5 may be displayed. Also,
The vehicle position on the road map display 1 is displayed on the current position display 6 as a coordinate value. The direction of the road map display 1 is displayed as a map direction display 7. The distance from the vehicle position 2 to the destination is displayed on the distance guidance display 8. When the vehicle position 2 approaches the guidance point 4, the intersection which is the guidance point is enlarged and displayed on a part of the screen as in the intersection enlarged display 9.

[0004] In the conventional navigation device, not only guidance using a screen as shown in FIG. 13 but also guidance providing voice is also used. Table 1 below shows an example of voice guidance for a normal guidance point 4 such as an intersection.

[0005]

[Table 1]

[0006] As shown in Table 2, voice guidance may be provided for a relay point or a destination point which is selected in advance on a route on which a vehicle travels and does not change the traveling route.

[0007]

[Table 2]

In the conventional navigation device, the vehicle position is detected by three-dimensional detection of the current position of the vehicle using satellite navigation using radio waves from artificial satellites by GPS, and vehicle speed information and the traveling direction of the vehicle. The information is used in combination with the detection of relative vehicle position using self-contained navigation, or the road shape of the road map data is compared with the shape of the route actually traveled by the vehicle, and the shapes match. The accuracy of vehicle position detection is ensured by using a method such as map matching or pattern matching that determines that a vehicle travels on a high-grade road.

[0009]

In a conventional navigation device, a vehicle position is absolutely detected by a GPS or the like, or a current vehicle position is relatively detected from a known vehicle position by using self-contained navigation. However, such detection accuracy is not always sufficient when compared with the width and density of the road on which the vehicle runs. If the road density is high,
It may not be possible to determine which road the vehicle is traveling on. Regarding GPS, a system called differential GPS has also been proposed. In this system, a moving vehicle receives a radio wave from a GPS artificial satellite and also receives a radio wave at a position fixed on the ground. At the fixed position, an error of the GPS radio wave is detected. Even if the vehicle receives the detection result almost simultaneously with the GPS radio wave and adopts a method of increasing the detection accuracy of the vehicle position, an error of about several meters may remain.

[0010] Even if the vehicle position can be accurately detected, the road map data represents a two-dimensional map based on aerial photographs, and is used under an overpass where a plurality of roads exist at the same latitude / longitude. In a place such as a highway, an elevated highway or a bypass road, the own vehicle position exists on a plurality of roads, and it becomes difficult to specify the road. In such a case, a road having a high degree of similarity is determined from a traveling locus to such a place by using a technique such as map matching or pattern matching. When such a traveling locus cannot be obtained or when the route is actually changed from one side to the other at the entrance to the elevated road, the user makes a judgment and manually positions the vehicle on the map. If not set, a normal vehicle position cannot be maintained, it is determined that the vehicle is traveling on an incorrect road, and route guidance and the like are not normally performed.

Further, even if the intersection enlarged display 9 is performed at the guide point 4 as shown in FIG. 13, on a road having a plurality of lanes, a correct lane is selected before reaching the guide point 4. There is a need. Conventional navigation devices do not have the accuracy required to detect the current position including the lane, so when performing route guidance including lane change at a guide point, regardless of which lane the vehicle is traveling in The same route guidance is performed.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a vehicle position detecting device and method capable of recognizing a lane of a road on which a vehicle travels, detecting a vehicle position and performing route guidance in consideration of the lane, and a route guiding device. And a method.

[0013]

SUMMARY OF THE INVENTION The present invention relates to a vehicle position detecting device mounted on a vehicle running on a road and detecting a vehicle position on the road by referring to road map data stored in advance. The road map data includes related information including the number of lanes, and processes an image of the road surface on which the vehicle travels, and an image of the road surface captured by the imaging unit.
A lane recognition unit for recognizing the lane displayed on the road surface to determine the number of lanes; and comparing the number of lanes determined by the lane recognition unit with the related information included in the road map data. And a vehicle position determining means for determining a vehicle position.

According to the present invention, the road surface on which the vehicle travels is imaged by the imaging means, and the image of the road surface is processed by the lane recognition means to recognize the lane and determine the number of lanes. The means can make a road decision based on the number of lanes. Since the road map data includes related information including the number of lanes, by comparing with the number of lanes determined by the lane recognizing means, roads that do not match the number of lanes are excluded from targets for determining the vehicle position, and The vehicle position can be reliably determined on the road on which the vehicle is traveling.

Further, in the present invention, the lane recognizing means also determines the lane in which the vehicle is traveling when the lane number is determined to be plural.

According to the present invention, since the lane in which the vehicle is traveling is also determined by the lane recognition means, it is possible to easily detect which lane the vehicle is traveling.

In the present invention, the vehicle position determining means may include:
A road for which a vehicle position is determined is selected by map matching, and a road that is a candidate for map matching selection is determined based on a comparison between the number of lanes determined by the lane recognition means and related information included in the road map data. , Sorting.

According to the present invention, the navigation for determining the vehicle position on the road by map matching is selected based on the number of lanes, so that the road determined by map matching is the road that is actually running. Possibilities can be increased.

Further, the present invention relates to a vehicle position detecting method for detecting a vehicle position on a road by referring to road map data stored in advance for a vehicle traveling on a road. The related information including the number is included, the road surface on which the vehicle travels is imaged, the image of the road surface to be imaged is processed, the lane is recognized and the number of lanes is determined, and the determined number of lanes is represented by road map data. A vehicle position is determined on a road having the same number of lanes as compared with related information included in the vehicle position detection method.

According to the present invention, the lane is recognized based on the image of the road on which the vehicle is traveling, the number of lanes is determined, and the number of lanes included in the road map data stored in advance is determined. Since the vehicle position is determined on a road having the same number of lanes as compared with related data including the vehicle number, the detection accuracy of the vehicle position can be improved.

Further, the present invention relates to a route guidance apparatus mounted on a vehicle traveling on a road and guiding the vehicle according to a preset route by referring to road map data stored in advance. The image processing device processes the road surface image captured by the image capturing means for capturing the road surface on which the vehicle travels, and determines the number of lanes by recognizing the lane displayed on the road surface. Lane recognizing means, vehicle position determining means for comparing the number of lanes determined by the lane recognizing means with related information included in road map data, and determining a vehicle position on a road having the same number of lanes; Route guidance means for guiding the traveling route of the vehicle based on the vehicle position determined by the means so as to follow the preset route.

According to the present invention, the image taken by the image pickup means of the road on which the vehicle travels is processed by the lane recognition means, so that the lane is recognized and the number of lanes is determined. The vehicle position detection means determines the road on which the vehicle travels based on the number of lanes, and the route guidance means performs route guidance assuming that the vehicle position exists on the determined road, so that the vehicle is actually traveling. Appropriate route guidance based on roads can be performed.

In the present invention, when the lane recognition means determines that the number of lanes is plural, the lane recognition means also determines the lane in which the vehicle is traveling, and the route guide means determines the lane determined by the lane recognition means. The feature is that route guidance is performed in consideration of the above.

According to the present invention, the route guidance means performs route guidance in consideration of the lane determined by the lane recognition means. Therefore, even if the vehicle is traveling on a road having a plurality of lanes, the lane change is performed. And appropriate route guidance including

The present invention further provides a route guidance method for guiding a vehicle traveling on a road according to a preset route with reference to road map data stored in advance. Include related information including, image the road surface on which the vehicle is traveling, process the image of the imaged road surface, recognize the lanes displayed on the road surface, determine the number of lanes, determine the number of lanes to be determined In comparison with the related information included in the road map data, a vehicle position is determined on a road having the same number of lanes, and based on the determined vehicle position, the traveling route of the vehicle follows the preset route. This is a route guidance method characterized by providing guidance.

According to the present invention, since the number of lanes is determined by imaging the road surface on which the vehicle travels, and the traveling route is guided based on the number of lanes, appropriate route guidance suitable for the actual traveling state of the vehicle is provided. It can be carried out.

[0027]

FIG. 1 shows a schematic electrical configuration of an embodiment of the present invention. The vehicle position detection device or the route guidance device according to the present embodiment includes a navigation device 10 and a traveling environment recognition device 11. The navigation device 10 includes a vehicle position detection device 12, a map display / route guidance device 13, and a display output device 14. The vehicle position detecting device 12 is configured to generate a map CD-R based on signals from various sensors for detecting GPS, gyro, vehicle speed, and the like.
Referring to the road map data stored in the OM in advance,
The vehicle position is detected as a position on the road. The map display / route guide device 13 reads map data from a map CD-ROM by inputting to operation keys or the like, and outputs the read data to a display output device 14.
To provide route guidance while displaying images. The video camera 15 of the traveling environment recognition device 11 captures an image of a road surface on which the vehicle travels, and inputs a video signal representing an image to the image recognition device 16. The image recognition device 16 recognizes lanes by image processing and determines the number of lanes.

FIG. 2 shows an algorithm of image processing in the image recognition device 16 of FIG. In this algorithm, processing as the white line recognition means 20 and the preceding vehicle recognition means 21 is mainly performed. The image captured by the video camera 15 in FIG. 1 is subjected to edge extraction processing 22 as image input. When an edge is extracted, white line recognition is performed through candidate point extraction processing 23, short-distance tracking processing 24, and long-distance tracking processing 25 that constitute the white line recognition means 20. Image input is performed from the time series smoothing processing 26 to the edge extraction processing 27
, And is also given to the preceding vehicle recognition means 21. In the preceding vehicle recognizing means 21, processing is performed by a bottom extraction processing 28, a search area setting processing 29, a projection processing 30, and a vehicle position determination processing 31. The recognition results of the white line recognition means 20 and the preceding vehicle recognition means 21 are given to a distance calculation process 32,
The inter-vehicle distance to the preceding vehicle is calculated, and an image in which the recognition result is written in the recognition result writing process 33 is output.
The navigation device 10 shown in FIG.
In the stage of candidate point extraction processing 23, the recognition result of the number of lanes,
The determination result of the lane in which the vehicle is traveling is given.

FIG. 3 shows a procedure of vehicle position detection performed by the vehicle position detection device 12 of FIG. Step a1
, The vehicle position is detected based on various sensors in step a2. In step a3, an image of a road surface is captured by the video camera 15. In step a5, lane recognition is performed by the white line recognition means 20 shown in FIG.
Is also determined. In step a6, the vehicle position detecting device 12 searches for a road having the same number of lanes while referring to the map CD-ROM. In step a7, the vehicle position is determined on the searched road. Thus, even when a plurality of roads intersect at the same latitude / longitude, if the number of lanes is different, it is possible to accurately determine on which road the vehicle is located. In step a8,
One vehicle position determination process ends. The processing from step a1 to step a8 is repeated at regular time intervals, and the change in the position of the vehicle traveling on the road can be tracked correctly.

FIG. 4 shows a basic concept for improving the detection accuracy by correcting the vehicle position on the road as another embodiment of the present invention. In the present embodiment, as shown in FIG. 4A, the vehicle position determined by processing the output from a sensor or the like, the distance L to the road, the vehicle position detection device 12 in FIG. Based on the traveling direction d and the direction D of the road, as shown in FIGS. 4B and 4C, the direction similarity f and the distance similarity g are calculated, respectively, and the direction similarity f and the distance similarity g are calculated. Are added with predetermined weighting factors F and G to calculate the similarity E.

FIG. 5 shows a processing procedure in an embodiment to which the concept shown in FIG. 4 is applied. The process starts from step b1, and in step b2, a road around the vehicle position is extracted. In step b3, a candidate is selected as a candidate only when the number of lanes on the peripheral road matches the number of lanes determined by the traveling environment recognition device 11 in FIG. 1, and the number N of peripheral links selected as a candidate among the peripheral links is determined. Set. In step b4, the parameter n is initialized to 0, and the maximum similarity Emax is set to the minimum similarity Emin.
Equal to At step b5, it is determined whether or not the value of n is smaller than the number N of peripheral links. If it is determined to be smaller, a similarity calculation is performed in step b6. In step b7, it is determined whether or not the calculated similarity E is greater than Emax. If it is larger, step b8
Then, the value of the similarity E calculated for Emax is substituted, and the current position H is corrected to the position h on the n-th link. When the condition is not satisfied in step b7, or in step b8
After, the parameter n is increased by 1 in step b9, and the process returns to step b5. When the condition is not satisfied in step b5, that is, when the value of the parameter n has reached N, it is determined in step b10 whether Emax and Emin are different. When they are different,
In step b11, the vehicle position is corrected to the value of H. When the condition is not satisfied in step b10 or in step b11
After, the process ends in step b12.

FIG. 6 shows a schematic system configuration of the driving environment recognition device 11 of FIG. The vehicle 40 is provided with a video camera 15 above the driver's seat, and the display output device 14 is provided near the driver's seat. The navigation device 10, the image recognition device 16, and the radar device 41 are also mounted in the vehicle 40.

FIG. 7 shows an example of a recognition result by the image recognition device 16. In the example of these recognition results, a black line is displayed over the image recognized as a white line. The video camera 15 is mounted so as to cover the driving field of view of the driver of the vehicle 40. To enhance recognition at night, a near-infrared CCD camera with a built-in image intensifier (optical amplifier) can be used for night.

In FIG. 7, a white line on the road is recognized as a lane mark, and the result is indicated by a thick black line. The preceding vehicle on the traveling lane of the own vehicle is indicated by a white rectangle. The judgment of the vehicle in front is made including the output from the radar. FIG. 7 (a)
Shows the case where there is a shadow generated from the roadside or a vehicle in an adjacent lane, and it can be seen that the white line is correctly recognized even in the presence of these. FIG. 7B shows a case where the traveling lane has a curve with a small radius of curvature. In this case also, it can be seen that the lane is correctly recognized. The radius of curvature of this curve is the level of the minimum value on the expressway determined by the example of the road structure, and is most difficult to recognize. Therefore, lanes can be correctly recognized on all expressways.

FIG. 8 shows a data structure of roads stored in the map CD-ROM in advance. Road map data 50
Is composed of a header 51, a node table 52, a link table 53, and an adjacent node table 54. The header 51 includes data such as “total data size” and “latitude / longitude”. The node table 52 includes a plurality of nodes, and each node includes data such as “node coordinates” and “number of branches”. The link table 53 includes a plurality of links, and each link includes “road type”,
Data such as "number of lanes" and "width" are included. The adjacent node table 54 includes a list of adjacent nodes. The road data is stored as road network information based on nodes representing such specific points and links connecting the nodes. In the present embodiment, the data of the “lane number” stored in each link is used to make a comparison with the lane number determined by the traveling environment recognition device 11.

In the image processing shown in FIG.
0 and the preceding vehicle recognizing means 21 basically perform the processing in parallel. However, since the process of recognizing the preceding vehicle is performed in the traveling lane in which the own vehicle travels, the process of the long distance side tracking process 25 in the white line recognition means 20 is completed, and the vehicle in front is recognized. The lowest extraction processing 28 of the recognition means 21 is performed. The image input is given to the edge extraction processing 22 and the time-series smoothing processing 26 as a monochrome signal. When performing recognition processing from an image, for example, a white line cannot always be defined as a “white portion on a screen” due to the influence of a shadow or the like. If the brightness of the white line changes depending on such a position in the screen, it is difficult to perform a recognition process characterized by shading. For recognition processing outdoors or the like, recognition processing using an edge image that is resistant to changes in brightness is effective.

In the edge extraction processes 22 and 27, a portion where the amount of change in the grayscale value is large in the image, that is, an outline is extracted, and an edge image which is an image composed only of the outline is obtained. FIG. 9A shows a normal image, and FIG. 9B shows an edge image.

As the edge, a positive edge and a negative edge can be extracted depending on the direction of the density change. There are also horizontal and vertical directions for extracting edges.
An edge image that is generally used does not consider the directionality of the edge, and includes two parts, an edge part and a non-edge part.
It is classified into types and binarized. Since the edge image as shown in FIG. 10 includes only the information on the outline, if a plurality of edges appear at the time of recognition of the white line, it becomes impossible to determine which side of the edge is the white line on the left or right. Therefore, an algorithm for performing recognition based on a nine-valued image classified into nine types based on the extraction direction and the sign of the edge without using the binarization is used. By using the information on the direction of the density change, the recognition accuracy can be improved.

Although it is effective to recognize an edge as a feature in the recognition of a vehicle ahead, if there is a mixture of shadow edges from a roadside object, it may be erroneously recognized. Therefore, as a method of reducing the influence of the shadow from the roadside object, a process of smoothing the image in the time direction is performed as the time-series smoothing process 26 in FIG. By the time-series smoothing processing 26, the shadow of an object moving at high speed on the screen, such as the shadow of a roadside object, fades, and no edge is extracted. On the other hand, an edge can be extracted from an object such as a preceding vehicle that is following and traveling relatively little on the screen.

FIG. 10A shows the concept of white line recognition. In the image captured by the video camera 15 mounted on the vehicle 40, the farther away the scenery other than the road surface occupies, the more difficult it is to determine whether or not it is a white line. Therefore, the white line recognition means 20 divides the screen area into a short distance side and a long distance side, and determines a lane in which the vehicle travels. In the candidate point extraction process 23, a plurality of candidate points that satisfy the directionality of the two edges and the white line width shown in the example of the road structure in the image area on the short distance side are extracted. In the short distance side tracking processing 24, the white line candidate points are weighted in accordance with the lane width condition shown in the road structure example, the white line candidate points are connected based on their linearity, and the white line position on the short distance side is determined. . Long distance side tracking processing 25
Then, the edge is traced along the outline of the white line from the short distance side to recognize the white line. Further, in order to cope with the case where the white line is a broken line that is not continuous, a tracking process of the white line based on the distance or jump is also performed. As described above, by utilizing the knowledge of the direction of the edge, the width of the white line, the width of the lane, and the like, it is possible to reduce the influence of stains and shadows on the road and improve the detection accuracy.

FIG. 10B shows the concept of recognition of a preceding vehicle. The front vehicle recognizing means 21 searches for a position in the traveling lane of the own vehicle where a long edge in the horizontal direction having a certain length or more appears by the bottom extraction processing 28. Since a large number of edges appear at the rear of the front vehicle, the distribution state of the edges is used as a recognition feature of the front vehicle. In order to be distinguished from edges other than the vehicle, the horizontal edge is extracted as the lowermost part of the front vehicle only when a plurality of horizontal edges appear vertically adjacent to each other. In the search area setting processing 29, a horizontally long search area is set which is set at the height of the front vehicle from the lowest position of the front vehicle extracted by the lowest extraction processing. In the projection processing 30, the projection of the edge in the search area is obtained. Shading of pixels along a vertical straight line is sequentially added, and the total is defined as a projection. In the vehicle position determination processing 31, it is determined that the position where the projection result is maximum in the width direction of the search area of the front vehicle is the position of the front vehicle based on the projection result.

FIG. 11 shows an example of a result display in which the captured image of the video camera 15 displays the recognition result of the white line and the measurement result of the inter-vehicle distance to the preceding vehicle. FIG.
Are converted to coordinates and displayed two-dimensionally.

When a road has a plurality of lanes, the number of white lines is increased by one from the number of lanes. Therefore, by counting the number of white lines, the number of lanes can be determined.

Further, it is possible to easily determine which lane of a plurality of lanes the vehicle is traveling, and if route guidance including lane information is performed, a right or left turn can be determined in advance. Guidance to an appropriate lane and appropriate guidance in consideration of the traveling lane can provide more effective route guidance.

[0045]

As described above, according to the present invention, a road surface is imaged, a lane is recognized by image processing, the number of lanes is determined, and the lane number is stored as related information of prestored road map data. Since the road having the same number of lanes is determined as the road on which the current position of the vehicle exists as compared with the number of lanes, the vehicle position can be correctly determined on the road on which the vehicle is traveling. Even if there are a plurality of roads at the same latitude / longitude, if the number of lanes is different, it can be easily determined and the vehicle position can be determined on the correct road.

Further, according to the present invention, it is possible to determine the lane from the image of the road and to determine which of the plurality of lanes the vehicle is traveling, so that the detection accuracy of the vehicle position can be improved. Can be further enhanced.

According to the present invention, candidate roads are selected based on the number of lanes when a vehicle position is set on a road by map matching, so that the road selected by map matching is actually running. , And the time required for processing can be reduced.

Further, according to the present invention, the number of lanes is determined from the image of the road, the road on which the vehicle is traveling is selected based on the number of lanes, and the position of the vehicle is determined on the selected road. An accurate vehicle position can be detected.

Further, according to the present invention, a road is selected in consideration of the number of lanes recognized by image processing, so that it is possible to increase the possibility of selecting a road that is actually traveling, and to guide an appropriate traveling route. It can be performed.

Further, according to the present invention, when a plurality of lanes are present on the road on which the vehicle is traveling, the lane in which the vehicle is traveling can also be determined. It is possible to provide guidance such as lane change at an appropriate timing.

Further, according to the present invention, the current position of the vehicle is determined on the road determined based on the number of lanes, and the vehicle follows a route set in advance based on the determined vehicle position. By setting on the correct road, reliable route guidance can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic electrical configuration of an embodiment of the present invention.

FIG. 2 is a block diagram showing an algorithm of image processing in the image recognition device 16 of FIG.

FIG. 3 is a flowchart illustrating a processing procedure of the embodiment of FIG. 1;

FIG. 4 is a diagram illustrating a concept of a similarity used for correcting a vehicle position according to another embodiment of the present invention.

FIG. 5 is a flowchart showing a procedure for correcting a vehicle position using the similarity of FIG. 4;

FIG. 6 is a simplified perspective view showing a state in which a device according to still another embodiment of the present invention is mounted on a vehicle.

FIG. 7 is a diagram illustrating a recognition result of the embodiment of FIG. 6;

FIG. 8 is a diagram showing an example of the configuration of road map data including the number of lanes.

FIG. 9 is a diagram illustrating an imaging screen of the video camera 15 to be subjected to image processing in the embodiment of FIG. 7 and an edge image obtained by performing edge extraction processing.

FIG. 10 is a diagram illustrating the concept of white line recognition and forward vehicle recognition in the image processing of FIG. 2;

11 is a diagram illustrating another example of the recognition result of the embodiment in FIG. 7;

12 is a diagram showing the recognition result of FIG. 11 after coordinate transformation.

FIG. 13 is a diagram showing an example of guidance display in a conventional navigation device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Navigation apparatus 11 Driving environment recognition apparatus 12 Vehicle position detection apparatus 13 Map display / route guidance apparatus 14 Display output apparatus 15 Video camera 16 Image recognition apparatus 20 White line recognition means 22 Edge extraction processing 23 Candidate point extraction processing 24 Short distance side tracking processing 25 Long-distance tracking processing 40 Vehicle 50 Road map data

Claims (7)

[Claims] 1. A vehicle position detecting device mounted on a vehicle traveling on a road and detecting a vehicle position on the road with reference to road map data stored in advance, wherein the road map data includes the number of lanes. And imaging means for imaging the road surface on which the vehicle travels, and a lane for processing the image of the road surface imaged by the imaging means and recognizing the lane displayed on the road surface to determine the number of lanes. Recognition means, and vehicle position determination means for comparing the number of lanes determined by the lane recognition means with related information included in the road map data and determining a vehicle position on a road having the same number of lanes. Vehicle position detecting device. 2. The vehicle position detecting device according to claim 1, wherein the lane recognition unit also determines the lane in which the vehicle is traveling when the number of lanes is determined to be plural. 3. The vehicle position determining means selects a road for which a vehicle position is determined by map matching, and selects a road that is a candidate for the map matching by the number of lanes determined by the lane recognizing means and the road map. 3. The vehicle position detecting device according to claim 1, wherein the sorting is performed based on comparison with related information included in the data. 4. A vehicle position detecting method for detecting a vehicle position on a road by referring to road map data stored in advance for a vehicle traveling on a road, wherein the road map data includes the number of lanes. The related information is included, the road surface on which the vehicle travels is imaged, the image of the road surface is processed, the lanes displayed on the road surface are recognized and the number of lanes is determined. A vehicle position detection method, comprising: comparing vehicle information with related information included in road map data to determine a vehicle position on a road having the same number of lanes. 5. A route guidance apparatus mounted on a vehicle running on a road and guiding according to a preset route by referring to road map data stored in advance, wherein the road map data includes the number of lanes. Image capturing means for capturing the road surface on which the vehicle travels, including related information, and lane recognition means for processing the image of the road surface captured by the image capturing means and recognizing the lanes displayed on the road surface to determine the number of lanes And the vehicle position determining means for comparing the number of lanes determined by the lane recognizing means with related information included in the road map data and determining the vehicle position on a road having the same number of lanes; Route guidance means for guiding the vehicle to follow the preset route based on the position of the vehicle. 6. The lane recognizing means also determines a lane in which the vehicle is traveling when the number of lanes is determined to be plural, and the route guiding means considers the lane determined by the lane recognizing means. The route guidance device according to claim 5, wherein the route guidance is performed. 7. A route guidance method for guiding a vehicle traveling on a road according to a preset route by referring to road map data stored in advance, wherein the road map data includes the number of lanes. The road surface on which the vehicle is traveling is imaged, the image of the road surface is processed, the lanes displayed on the road surface are recognized and the number of lanes is determined. Comparing the relevant information included in the data, determining a vehicle position on a road having the same number of lanes, and guiding the vehicle so that the traveling route of the vehicle follows the preset route based on the determined vehicle position. A route guidance method characterized by the following.