JP2003227725A - On-vehicle navigation system, navigation method, and program for navigation - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide navigation technology, namely, an on-vehicle navigation system, a navigation method, and a program for navigation by which excellent matching accuracy can be realized, by discriminating the type of a road photographed with a camera as an ordinary road, superhighway, etc., from the special feature of the road. <P>SOLUTION: The on-vehicle navigation system discriminates the traveling road of its own vehicle based on map data, the detected location of its own vehicle, and the special feature of the photographed road, by performing pattern matching by means of a pattern matching section 1A and map matching by means of a map matching section 1B and shows the location of its own vehicle on the discriminated road to the user in a displaying section B. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a navigation technique for realizing excellent matching accuracy by discriminating a road type such as a general road or a highway from the characteristics of a road photographed by a camera, that is, an in-vehicle navigation system and The present invention relates to a navigation method and a navigation program.

[0002]

2. Description of the Related Art In recent years, the technology of vehicle-mounted navigation systems has been rapidly advancing along with the spread of automobiles and the progress of information processing technology. The in-vehicle navigation system calculates the optimal route to the specified destination based on the road map data prepared in advance, and displays the vehicle position and the surrounding map that are measured and estimated in real time, while displaying left and right turns. Navigation is performed by synthetic voice or screen display.

For estimating the vehicle position in such navigation, absolute position detecting means such as GPS (Global Positioning System) and relative position detecting means such as azimuth sensor and distance sensor are used. However, the own vehicle position calculated and estimated in this way may have an error in itself and an error in the road map data itself presented to the user, and thus may be inconsistent with the actual traveling state.

As a countermeasure against such a case, there is used a process for determining which road on the map the estimated position of the own vehicle corresponds, and this process is called map matching. This map matching is performed by comparing the traveling locus of the own vehicle with the shape of the road around the estimated position of the own vehicle on the map.
Here, the running locus of the own vehicle is obtained by connecting the coordinates of each position estimated at regular time intervals with a line segment. For comparison with a line similarly represented by a line or a line segment, for example, two lines are used. Use cross-correlation of figures.

However, when such a comparison process is performed using the CPU of the computer, the length of the running locus of the own vehicle is limited due to restrictions such as memory capacity, calculation time, and processing capacity of the CPU. This is also because the error of the sensor accumulates by the length of the locus and the comparison is not performed well, so that it is suppressed. Therefore, the length of the road to be compared corresponds to the length of the trajectory (or the length proportional to it).

[0006] For example, when map matching is performed for the vehicle position Pa point in FIG. 8 estimated by GPS or the like,
If the range of the estimation error of the sensor is set to a broken rectangle, the roads R1 and R2 included therein are candidates for map matching, and the road R2 is the most suitable road in consideration of the direction of the own vehicle. In this case, while shifting the coordinates of each point on the road R2 along the traveling locus Po → Pa and calculating some cross-correlations such as relative angles with the traveling locus, the shift amount with the highest correlation is the road. It is the vehicle position at R2.

[0007]

However, as shown in FIG. 9, a plurality of roads R having similar features such as shape and direction are used.
When 2 and R3 are close to each other within the error range of the sensor, there is a problem that the conclusion of the map matching cannot be specified as one by only the above-described azimuth comparison and shape comparison.

For example, when a highway and a general road are running in parallel, or when the highway is running on the general road, the conventional map matching cannot determine the difference. There was a frequent need to erroneously perform matching on a road or from an expressway to a general road, and there was a long-awaited solution.

The present invention has been proposed to solve the above-mentioned problems of the prior art, and its purpose is to discriminate the road type such as a general road or a highway from the characteristics of the road photographed by a camera. By doing so, it is to provide a navigation technology that realizes excellent matching accuracy, that is, an in-vehicle navigation system, a navigation method, and a navigation program.

[0010]

In order to achieve the above object, an in-vehicle navigation system according to claim 1 is a navigation device for navigating a vehicle, and includes map data including information on where and what road is located. In advance, a means for detecting the vehicle position of the vehicle, an image capturing means for capturing a feature of a road around the vehicle, the map data, the detected vehicle position, and Based on the characteristics of the road, means for determining which road the vehicle is on by map matching and means for displaying the position of the vehicle on the determined road are provided.

The navigation method of claim 8 is obtained by considering the invention of claim 1 as a method. In the navigation method for navigating a vehicle,
Map data including information on where and where the road is is stored in advance, the position of the vehicle of the vehicle is detected, a feature of the road around the vehicle is photographed, and the map data and the detected road are detected. Based on the vehicle position and the feature of the photographed road, which road the own vehicle is on is determined by map matching, and the own vehicle position on the determined road is displayed.

The navigation program of claim 12 is a computer program of the inventions of claims 1 and 8, and is a computer in which map data including information about where and what road is stored in advance. By controlling, in a navigation program for navigating a vehicle, the program causes the computer to detect the position of the vehicle of the vehicle, image the features of the road around the vehicle, and the map data, Based on the detected position of the own vehicle and the feature of the photographed road, the map matching is performed to determine which road the own vehicle is on, and the position of the own vehicle on the determined road is displayed. To do.

In these aspects, the type of road being run is discriminated from the characteristics of the road such as the type of white line based on the image data captured by a video camera or the like, and it is used in combination with other matching conditions so as to approach each other. The erroneous matching such as the difference between the normal road and the expressway is prevented, and the navigation accuracy is improved.

According to a second aspect of the present invention, in the vehicle-mounted navigation system according to the first aspect, the feature of the road is a paint pattern on the road surface.

According to a ninth aspect of the present invention, in the navigation method according to the eighth aspect, the feature of the road is a paint pattern on the road surface.

In these modes, the road type is discriminated based on the road surface paint pattern such as a white line pattern, the number of lines drawn in parallel, the presence or absence of a pedestrian crossing or a stop line. It is possible to perform map matching with high accuracy by considering the scale of roads.

According to a third aspect of the present invention, in the vehicle-mounted navigation system according to the first or second aspect, the imaging means is an infrared rear confirmation camera.

According to a tenth aspect of the invention, in the navigation method according to the eighth or ninth aspect, the photographing is performed by using an infrared rear confirmation camera.

In these modes, since the road information is captured by the infrared rear confirmation camera, the excellent accuracy of map matching is effectively maintained even at night, and a new camera is not required.

According to a fourth aspect of the invention, in the vehicle-mounted navigation system according to any of the first to third aspects, regarding the feature of the road, the feature of the white line on the road surface is identified to identify the feature of the road on which the vehicle is traveling. It is characterized in that it is configured to determine the type.

The invention according to claim 11 is the navigation method according to any one of claims 8 to 10, wherein the type of road on which the vehicle is traveling is identified by identifying the feature of the white line on the road surface with respect to the feature of the road. It is characterized by determining.

In these modes, a small processing load can be obtained by identifying patterns such as continuous and intermittent lines and features such as width and number of white lines on the road surface, and judging the type of road such as a highway or a general road based on the features. High-accuracy map matching is realized.

According to a fifth aspect of the invention, in the vehicle-mounted navigation system according to the fourth aspect, the features of the white line are a length and a ratio of a white line portion and a blank portion between the white line portions. .

According to a sixth aspect of the present invention, in the vehicle-mounted navigation system according to the fifth aspect, the length and the ratio are such that the total length of the white line portion and the blank portion is 20 meters per set, and the total length is 20 meters. When the ratio of the blank portion in the inside is within a predetermined allowable error range based on 60%, the candidate for map matching is configured to be limited to the expressway.

[0025] In these modes, for the white line intermittently drawn on the road surface, the total length and the length ratio of the white line portion and the blank portion are mechanically compared to a predetermined reference to mechanically drive the road. Since the highway can be identified, the processing load is reduced. The highway includes various roads having a predetermined structure, such as high-standard highways and other highways.

According to a seventh aspect of the present invention, in the vehicle-mounted navigation system according to any one of the first to sixth aspects, a plurality of comparison and contrast patterns under different photographing conditions are prepared for white lines that are intermittent on the road surface, and the photographed image is displayed. It is characterized in that at least one of binarization and contour extraction is performed, the comparison is made with each comparison pattern, and the type of road is judged based on the degree of coincidence.

In this embodiment, a plurality of comparison / contrast patterns (templates) are prepared according to different photographing conditions such as the mounting angle and position of the camera, and compared with the photographed images binarized or contour-extracted to determine the road type based on the degree of coincidence. to decide. Therefore, complicated and sophisticated processing such as straight line and shape recognition from a photographed image is not necessary, and smooth navigation processing can be performed with limited CPU processing capacity.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention (hereinafter referred to as "embodiments") will be specifically described with reference to the drawings. It should be noted that the present embodiment is realized by controlling a computer with software. However, in this case, the implementation modes of hardware and software can be changed in various ways. Therefore, in the following description, the present invention and the embodiment will be described. Virtual circuit blocks that realize each function are used.

[1. Configuration] The present embodiment relates to an in-vehicle navigation system (hereinafter referred to as "this system") for performing vehicle navigation and a navigation method executed on this system, and can be understood as a navigation program.

First, as shown in the functional block diagram of FIG. 1, this system comprises a main body A, a display B and an infrared rear confirmation camera C, and the following elements shown in FIG. That is, in the main body A, 1 is a main CPU that controls the entire system, 2 is a storage unit such as DRAM that provides a system work area and program storage area, and 3 is a display control that draws a map or menu. Parts 4 are peripheral circuits of the CPU such as an interface circuit (I / F) of each device.

Further, 5 is a CD-ROM or a DVD-RO
It is a storage medium such as an M drive or a hard disk drive, and includes a road map database. This road map database is a means for preliminarily storing map data for route search and map display, such as network structure of where and what kind of road is there, targets such as buildings, and network type or relational type. It is a database.

Further, 6 is a relative azimuth detecting section using a gyro sensor or the like, 7 is an absolute position detecting section such as GPS,
These are means for detecting the vehicle position of the vehicle in real time. Further, 8 is a voice synthesis unit for performing voice guidance, and 9
Is an output unit such as an amplifier for driving a guide speaker (SP), 10 is a vehicle I / F unit for taking in vehicle information such as a vehicle speed signal and a back signal, and 11 is a V by FM multiplexing or a beacon.
It is a VICS information receiving unit that takes in ICS information.

In the display section B, 12 is a video (VIDEO) signal processing section for displaying a video signal on the display 13, 13 is a display such as an LCD or CRT for displaying maps and menus, and 14 is a touch panel. An input unit such as or an operation key, and 15 is a remote control light receiving unit for receiving a remote control signal.

Further, the infrared rear confirmation camera section C is installed in, for example, the tail of the vehicle, and 16 is a device for capturing a video signal, that is, a CCD and its optical system,
Reference numeral 17 is a VIDEO signal processing section, and in the main body section A, 18 is an A / D conversion section for the VIDEO signal. The CCD, the optical system 16 and the VIDEO signal processing unit 17 in the infrared rear confirmation camera section C constitute an image pickup means for photographing the feature of the road, and the feature of the road in this embodiment is a paint pattern on the road surface. Also,
It may be considered that the image pickup means includes the A / D conversion section 18 of the main body section A.

Further, the main CPU 1 operates as follows according to the program stored in the storage unit 2, and its function can be grasped separately for the pattern matching unit 1A and the map matching unit 1B shown in FIG. it can.

[2. Action] That is, in the present embodiment,
By the pattern matching performed by the pattern matching unit 1A and the map matching performed by the map matching unit 1B, the map data, the detected vehicle position, and
Based on the captured characteristics of the road, it is determined which road the vehicle is on, and the position of the vehicle on the determined road is displayed to the user on the display unit B.

[2-1. Pattern Matching] Of these, the pattern matching is performed by the pattern matching unit 1A.
This is a process of limiting the types of candidates for map matching from highways and general roads. In this pattern matching, the type of road on which the vehicle is traveling is determined by identifying the features of the white line on the road surface regarding the features of the road. In this case, the features of the white line are the white line part and the space between the white line parts. The length and the ratio with the part.

Specifically, regarding the length and the ratio, the total length of the white line part and the blank part is 20 meters per set, and the ratio of the blank part in the total length is 60%.
The target candidates for map matching are limited to the expressways when they are within the predetermined allowable error ranges with respect to each other.

As a more specific method of this pattern matching, a comparison and contrast pattern (called a template) under different shooting conditions for white lines that are intermittent on the road surface is used.
Is prepared, at least one of binarization and contour extraction of the photographed image is performed, and the images are compared with each template, and the type of road is determined based on the degree of coincidence.

That is, by utilizing the fact that the white line on the expressway is a repetition of white 8 m and blank 12 m, the above-mentioned 8 is used as a comparison pattern on the storage medium 5 such as a CD-ROM.
The figures when m and 12 m are taken in by the camera are prepared as several types of templates depending on the appearance depending on the mounting angle and position of the camera.

Here, the processing procedure of pattern matching is shown in the flowchart of FIG. That is, the video signal behind the vehicle is fetched from the VIDEO signal processing unit 17 of the infrared rear confirmation camera unit C (step 01). The video signal thus captured is converted into a digital signal by the A / D converter 18 of the main body A.

Here, the captured signal usually contains unnecessary information such as the white line, vehicles running around the road, roadside trees, lighting towers, etc. due to the nature of the rear confirmation camera, but the position of the white line from the running state. Are likely to concentrate in certain areas, such as the bottom center. Therefore, the main CPU 1
Only the periphery of the white line is cut out, and the contour is extracted by binarizing 0 or 1 (step 02). An example of the signal cut out in this way is shown in FIG.

Then, the pattern matching section 1A sequentially repeats the shapes T of the templates 1 to N prepared in advance in several types and the contour E as illustrated in FIG. 3 (steps 03 to 07 in FIG. 2). Compare (Step 0
4) When it is determined that they match with each other based on a predetermined overlapping area as illustrated in FIG. 4 (step 0
5) Judge as an expressway and set a specific bit K of a predetermined register N used as a flag to 1 (step 06).

On the other hand, as shown in FIG. 5, when the pattern matching results do not match, it is determined that the road is not a highway, and a bit K of a predetermined register N used as a flag is set to 0 (step 08). ). Ie 2
When two figures are superposed and compared, it can be determined that their lengths and ratios are different. Therefore, in the case of such a mismatch, the bit K of the register N becomes 0.

[2-2. Map matching] On the other hand, by time sharing (time division multiplex) or the like, in addition to the above-described pattern matching, processing such as drawing of a peripheral map and updating of the vehicle position is performed, and map matching is also performed by the map matching unit 1B. It is being appreciated. Then, the result of the pattern matching is passed to the map matching as follows.

That is, at the time of map matching, as shown in the flowchart of FIG. 6, following the trajectory drawing of the vehicle (step 11), the bit K of the register N is referred to, and if the value is 1 (step 12). ) The target road of the matching candidate is limited to the expressway (step 13) and the map matching is performed (step 14). In this case, for example, in the example of FIG. 7, the road R2 indicated by the broken line is excluded from the matching target roads, and the vehicle position is matched only with the road R3 (highway) which is the highway indicated by the thick solid line.

[3. [Effect] As described above, in the present embodiment, based on the image data captured by the video camera or the like, the type of road being driven is determined from the characteristics of the road such as the type of white line, and it is used in combination with other matching conditions. , It is possible to prevent erroneous matching such that a general road and an expressway approaching each other are mistaken, and navigation accuracy is improved.

In particular, in this embodiment, since the road type such as a highway or a general road is discriminated based on the paint pattern on the road surface, especially the intermittent pattern of white lines, it is possible to perform map matching with high accuracy.

Further, in the present embodiment, since the road information is captured by the infrared rear confirmation camera, the excellent accuracy of map matching is effectively maintained even at night, and a new camera is not required.

Further, in the present embodiment, regarding the white line intermittently drawn on the road surface, the total length and the length ratio of the white line portion and the blank portion are mechanically compared with a predetermined reference. Since it is possible to distinguish between general roads and highways, the processing load is reduced.

Further, in the present embodiment, the comparison and contrast pattern based on different photographing conditions such as the mounting angle and the position of the camera.
A plurality of (templates) are prepared, and the type of road is judged based on the degree of coincidence by comparing with the photographed image obtained by binarization or contour extraction.
Therefore, complicated and sophisticated processing such as straight line and shape recognition from a photographed image is not necessary, and smooth navigation processing can be performed with limited CPU processing capacity.

[4. Other Embodiments] The present invention is not limited to the above embodiments, and includes other embodiments as illustrated below. For example, an image of the front of the vehicle may be captured instead of the rear of the vehicle in order to determine the characteristics of the road around the vehicle. In addition, the characteristics of the road are not limited to intermittent white lines, crosswalks, stop lines, road signs, traffic lights,
You can freely set the speed limit display, presence of distance posts, etc., presence of elevated structures around the sidewalk and surroundings. Further, the judgment method is not limited to the degree of coincidence with the template, and the extracted contour may be theoretically analyzed.

[0053]

As described above, according to the present invention, a navigation technique for realizing excellent matching accuracy by discriminating a road type such as a general road or a highway from the characteristics of a road photographed by a camera, That is, since it is possible to provide an in-vehicle navigation system, a navigation method, and a navigation program, a safe and comfortable car life is realized.

[Brief description of drawings]

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

FIG. 2 is a flowchart showing a pattern matching processing procedure according to the embodiment of the present invention.

FIG. 3 is a diagram illustrating an example of a captured image on which processing such as contour extraction has been performed in the embodiment of the present invention.

FIG. 4 is a conceptual diagram showing matching between a contour and a template according to the embodiment of the present invention.

FIG. 5 is a conceptual diagram showing matching between a contour and a template according to the embodiment of the present invention.

FIG. 6 is a flowchart showing a processing procedure of map matching according to the embodiment of the present invention.

FIG. 7 is a conceptual diagram showing an example of map matching in the embodiment of the present invention.

FIG. 8 is a conceptual diagram showing an example of map matching.

FIG. 9 is a conceptual diagram showing an example of map matching.

[Explanation of symbols]

A ... Main body B ... Display 1 ... Main CPU 2 ... memory 3 ... Display control unit 4 ... CPU peripheral circuit 5 ... CD-ROM or DVD-ROM 6 ... Absolute position / direction detector 7 ... Direction detection unit 8 ... Voice section 9 ... Amplifier 10 ... Vehicle I / F 11 ... VICS information receiving unit 12 ... Signal processing unit 13 ... Indicator 14 ... Input section 15 ... Remote receiver 16 ... CCD and optical system 17 ... VIDEO signal processing unit 18 ... A / D converter

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI theme code (reference) G09B 29/10 G09B 29/10 AF term (reference) 2C032 HB02 HB22 HB23 HB24 HC08 HC15 HC23 HC31 HD16 HD30 2F029 AA02 AB01 AB07 AB12 AC02 AC08 AC09 AC18 AD02 5B057 AA16 BA02 CA12 CA16 DA06 DB02 DC01 5H180 AA01 BB13 CC02 CC04 CC24 EE18 FF07 FF12 FF13 FF22 FF25 FF33 FF37 5L096 BA04 CA02 FA03 HA07 JA03 JA09

Claims (12)

[Claims] 1. A navigation device for navigating a vehicle, means for pre-storing map data including information on where and what road is located, means for detecting a vehicle position of the vehicle, and the vehicle. Based on the image pickup means for photographing the characteristics of the surrounding road, the map data, the detected vehicle position, and the photographed road characteristics, it is determined by map matching which road the vehicle is on. An in-vehicle navigation system, comprising: a means for performing the operation and a means for displaying the position of the vehicle on the determined road. 2. The in-vehicle navigation system according to claim 1, wherein the feature of the road is a paint pattern on the road surface. 3. The vehicle-mounted navigation system according to claim 1, wherein the image pickup means is an infrared rear confirmation camera. 4. With respect to the feature of the road, the type of the road on which the vehicle is traveling is determined by identifying the feature of the white line on the road surface.
The in-vehicle navigation system according to any one of 1 to 3. 5. The in-vehicle navigation system according to claim 4, wherein the characteristic of the white line is a length and a ratio of a white line portion and a blank portion between the white line portions. 6. Regarding the length and the ratio, the total length of the white line part and the blank part is one set 20.
M, the ratio of the blank portion in the total length is 60%, and is configured to limit the candidate candidates for map matching to highways when they are within the respective allowable error ranges. The in-vehicle navigation system according to claim 5. 7. A plurality of comparison and contrast patterns under different photographing conditions are prepared for intermittent white lines on the road surface, and the photographed image is subjected to at least one of binarization and contour extraction, and each of the comparison and contrast patterns. The vehicle-mounted navigation system according to claim 1, wherein the vehicle-mounted navigation system is configured to determine the type of road based on the degree of coincidence with the above. 8. A navigation method for navigating a vehicle, wherein map data including information on where and what road is present is stored in advance, a vehicle position of the vehicle is detected, and a road around the vehicle is detected. A feature is photographed, based on the map data, the detected vehicle position, and the photographed road feature, it is determined by map matching which road the vehicle is on. A navigation method characterized by displaying a vehicle position. 9. The navigation method according to claim 8, wherein the feature of the road is a paint pattern of a road surface. 10. The navigation method according to claim 8, wherein an infrared rear confirmation camera is used for the photographing. 11. The navigation according to claim 8, wherein the type of the road on which the vehicle is traveling is determined by identifying the feature of the white line on the road surface with respect to the feature of the road. Method. 12. A navigation program for navigating a vehicle by controlling a computer in which map data including information on where and what road is located is stored in advance, the program being stored in the computer, The vehicle position of the vehicle is detected, the features of the road around the vehicle are photographed, and the road on which the vehicle is based on the map data, the detected vehicle position, and the photographed road feature. A navigation program characterized by displaying whether or not the vehicle position on the road is judged by map matching.