JP2008249555A - Position-specifying device, position-specifying method, and position-specifying program - Google Patents

Abstract

An object of the present invention is to calculate a vehicle position with high accuracy.
A pseudo-range positioning unit 210 obtains a positioning position A by pseudo-range positioning. The white line data acquisition unit 221 acquires the three-dimensional coordinate data of each white line located around the positioning position A. The white line data imaging plane projection unit 222 projects each white line data on the camera imaging plane. The video white line extraction unit 223 extracts a white line from the camera video. The feature amount projection unit 224 calculates the feature amount of each imaging plane white line data and video white line data, and the feature amount correlation coefficient calculation unit 225 compares each imaging plane white line data with the video white line data, and a correlation coefficient determination unit. Reference numeral 226 specifies imaging surface white line data corresponding to the video white line data. The vehicle position calculation unit 227 applies the three-dimensional coordinates indicated by the white line data to the two-dimensional position of the white line in the camera image, and calculates the vehicle position based on the camera characteristics such as the focal length, the angle of view, and the installation angle. .
[Selection] Figure 3

Description

  The present invention relates to a vehicle position specifying device for accurately grasping the position of a vehicle from feature points in an image of a TV camera mounted on the vehicle, for example.

  Although the position of the vehicle can be grasped with a certain degree of accuracy by GPS (Global Positioning System), some supplementary means for obtaining positioning information is necessary under the condition that a positioning signal (also called a carrier wave) cannot be received from a GPS satellite. there were.

For example, gyro compensation and vehicle speed pulse complement are generally used as supplementary techniques, and positioning is performed by inertial navigation or dead reckoning (dead reckoning). However, there is a problem in securing the positioning accuracy by such a complementary means. For example, because the positioning accuracy decreases due to the effects of drift in the gyro measurement results (angular acceleration of azimuth, elevation, and rotation angle), it is necessary to use an expensive gyro to suppress the effects of drift as much as possible. there were.
JP 2006-023278 A JP 2006-47291 A

  Conventionally, gyroscopes and wheel rotation speed counters that measure angular acceleration of azimuth angle, elevation angle, and rotation angle are used as supplementary technologies for GPS positioning in locations where GPS positioning signals are difficult to receive by a GPS receiver mounted on a vehicle. Inertial navigation and dead reckoning using vehicle speed pulses for calculating vehicle speed based on the above have been performed. In this case, for example, the drift of the gyro becomes a problem, and there is a problem that the position accuracy gradually deteriorates in a situation where the GPS positioning signal cannot be received for a long time.

  In addition, the conventional device for detecting the position of the vehicle by an image can grasp the position in the lateral direction in the lane by detecting a white line by image processing, but the absolute position (in particular, the absolute direction of the traveling direction). Since the position) is not detected, the position of the vehicle in the vertical direction cannot be grasped in detail.

  An object of the present invention is to specify the position of a vehicle with high accuracy, for example.

  The position specifying device of the present invention includes a first positioning unit that measures a self-position, a camera that captures a specific direction when the first positioning unit performs positioning, and an imaging surface when the camera captures an image. An image processing unit that uses a CPU (Central Processing Unit) to identify a specific imaging feature reflected in the image in a two-dimensional image, and a feature that indicates a three-dimensional position of the feature and a three-dimensional shape of the feature A feature database that stores information; a feature information acquisition unit that acquires feature information of each feature located around the self-position based on the self-position measured by the first positioning unit from the feature database; The feature indicated by the feature information based on the position of the feature indicated by the feature information acquired by the feature information acquisition unit and the shape of the feature is used as a projected feature on the imaging surface of the camera using the CPU. projection And comparing the projected features projected by the feature projecting unit with the specific features identified by the image processing unit using the CPU on the imaging plane. A feature information specifying unit for specifying a projected feature corresponding to the object, a two-dimensional position on the imaging surface of the projection feature specified by the feature information specifying unit, and a projection feature specified by the feature information specifying unit The second positioning unit that measures the self-position using the CPU based on the three-dimensional position of the feature indicated by the feature information, and the self-position measured by the second positioning unit as a positioning result to the output device And a positioning result output unit for outputting.

  According to the present invention, for example, the image processing unit specifies a white line from the camera video, the feature information acquisition unit acquires the actual three-dimensional coordinates of the white line from the feature database, and the second positioning unit acquires the camera video. By applying the three-dimensional coordinates indicated by the feature database to the position of the white line, the self-position can be calculated with high accuracy based on the characteristics of the camera.

Embodiment 1 FIG.
FIG. 1 is a device configuration diagram relating to positioning of a vehicle 100 equipped with a vehicle position specifying device 200 according to the first embodiment.
The vehicle 100 is an example of a positioning target. For example, the vehicle position specifying device 200 may be mounted on a motorcycle that is a moving body that travels on a road to determine the traveling position of the motorcycle.

  In FIG. 1, a vehicle 100 that is a positioning target includes a TV camera 110 that captures a road ahead of the vehicle 100 (the traveling direction, the azimuth direction of the vehicle 100, or the rear of the vehicle 100), and a GPS. A GPS receiver 120 that receives a positioning signal (carrier wave) from a satellite by a GPS antenna 121 and performs GPS observation, a road database 190 that stores road information used in a car navigation system, and a TV camera 110 The position of the vehicle 100 is identified with high accuracy based on the measured road image and the positioning position based on the GPS observation amount (for example, pseudorange, carrier wave phase) of the GPS receiver 120 and the surrounding road information acquired from the road database 190. The vehicle position specifying device 200 is provided.

  It is assumed that the TV camera 110 is attached to the vehicle 100 so that the imaging direction coincides with the length direction (traveling direction) of the vehicle 100. That is, when the vehicle 100 is traveling toward true north, the TV camera 110 is located at a field angle range corresponding to the characteristics and parameters (for example, focal length) of the TV camera 110 with true north as the center. Can be imaged. The video captured by the TV camera 110 may be a moving image or a still image.

The road database 190 (an example of a feature database) stores three-dimensional road data (road information).
In particular, the road database 190 displays an edge portion or a point in order to indicate a real three-dimensional position and shape for road surface display such as a center line (continuous line, intermittent line), a stop line, a pedestrian crossing, and a deceleration display of a road. It is assumed that three-dimensional coordinate data indicating the position of the end point of the edge is stored. Hereinafter, a white line will be described as an example of the road surface display. Also, the three-dimensional coordinate data for the white line is used as white line data. The road database 190 that stores the white line data can be constructed by using, for example, the technique of Patent Document 2.

FIG. 2 is a diagram illustrating an example of hardware resources of the vehicle position specifying device 200 according to the first embodiment.
In FIG. 2, the vehicle position specifying device 200 includes a CPU 911 (also referred to as a central processing unit, a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, and a processor) that executes a program. The CPU 911 is connected to the ROM 913, the RAM 914, the communication board 915, and the magnetic disk device 920 via the bus 912, and controls these hardware devices. Instead of the magnetic disk device 920, another storage device (for example, a semiconductor memory such as a RAM or a flash memory) may be used.
The RAM 914 is an example of a volatile memory. The storage medium of the ROM 913 and the magnetic disk device 920 is an example of a nonvolatile memory. These are examples of a storage device, a storage device, or a storage unit. A storage device in which input data is stored is an example of an input device, an input device, or an input unit, and a storage device in which output data is stored is an example of an output device, an output device, or an output unit.
The communication board 915 is an example of an input / output device, an input / output device, or an input / output unit.

  The communication board 915 is wired or wirelessly connected to a communication network such as a LAN (Local Area Network), the Internet, a WAN (Wide Area Network) such as ISDN, and a telephone line.

  The magnetic disk device 920 stores an OS 921 (operating system), a program group 923, and a file group 924. The programs in the program group 923 are executed by the CPU 911 and the OS 921.

  The program group 923 stores a program for executing a function described as “˜unit” in the embodiment. The program is read and executed by the CPU 911.

In the file group 924, in the embodiment, result data such as “determination result”, “calculation result of”, “processing result of” when executing the function of “to part”, “to part” The data to be passed between programs that execute the function “,” other information, data, signal values, variable values, and parameters are stored as items “˜file” and “˜database”. Positioning results, white line data acquired from the road database 190, camera images captured by the TV camera 110, pseudoranges observed by the GPS receiver 120, carrier wave information, and the like are examples of what is included in the file group 924.
The “˜file” and “˜database” are stored in a recording medium such as a disk or a memory. Information, data, signal values, variable values, and parameters stored in a storage medium such as a disk or memory are read out to the main memory or cache memory by the CPU 911 via a read / write circuit, and extracted, searched, referenced, compared, and calculated. Used for CPU operations such as calculation, processing, output, printing, and display. Information, data, signal values, variable values, and parameters are temporarily stored in the main memory, cache memory, and buffer memory during the CPU operations of extraction, search, reference, comparison, operation, calculation, processing, output, printing, and display. Is remembered.
In addition, arrows in the flowcharts described in the embodiments mainly indicate input / output of data and signals, and the data and signal values are recorded in the memory of the RAM 914, the magnetic disk of the magnetic disk device 920, and other recording media. . Data and signal values are transmitted online via a bus 912, signal lines, cables, or other transmission media.

  In addition, what is described as “˜unit” in the embodiment may be “˜circuit”, “˜device”, “˜device”, and “˜step”, “˜procedure”, “˜”. Processing ". That is, what is described as “˜unit” may be realized by firmware stored in the ROM 913. Alternatively, it may be implemented only by software, or only by hardware such as elements, devices, substrates, and wirings, by a combination of software and hardware, or by a combination of firmware. Firmware and software are stored as programs on a magnetic disk or other recording medium. The program is read by the CPU 911 and executed by the CPU 911. That is, the position specifying program causes the computer to function as “to part”. Alternatively, the procedure or method of “to part” is executed by a computer.

FIG. 3 is a functional configuration diagram of the vehicle position specifying device 200 according to the first embodiment.
The functional configuration of the vehicle position specifying device 200 in the first embodiment will be described below with reference to FIG.

  The vehicle position specifying device 200 includes a pseudo distance positioning unit 210 (an example of a first positioning unit) that calculates an approximate position (positioning position A) of the vehicle 100 based on the pseudo distance observed by the GPS receiver 120, a pseudo distance Corresponding to the camera position based on the white line data (three-dimensional coordinates) of the white line located around the positioning position A calculated by the positioning unit 210 and the position of the white line in the camera image captured by the TV camera 110. An image positioning unit 220 that calculates a detailed position (positioning position B) of the vehicle 100, and a positioning result output that outputs a positioning result (positioning position B) to a car navigation system (not shown), an automatic driving control device (not shown), or the like. Unit 230.

The image positioning unit 220 includes the following functional configuration.
A white line data acquisition unit 221 (an example of a feature information acquisition unit) inputs a positioning position A, which is an approximate positioning result of the pseudorange positioning unit 210, and a three-dimensional white line positioned around the vehicle 100 based on the positioning position A. White line data indicating coordinates is acquired from the road database 190.
The white line data imaging plane projection unit 222 (an example of a feature projection unit) projects white line white line data positioned around the vehicle 100 onto the imaging plane of the TV camera 110, and a two-dimensional position (for example, coordinates) of the white line on the imaging plane. Imaging plane white line data (an example of a projected feature) indicating (pixel position).
A video white line extraction unit 223 (an example of an image processing unit) detects a white line appearing in a camera video captured by the TV camera 110, and determines the two-dimensional position of the white line in the camera video (corresponding to the imaging surface of the TV camera 110). The image white line data shown (an example of a specific imaging feature) is calculated.
The feature amount projection unit 224 is perpendicular to the horizontal feature amount on the imaging surface of the TV camera 110 with respect to the imaging surface white line data calculated by the white line data imaging surface projection unit 222 and the video white line data calculated by the video white line extraction unit 223. The direction feature amount is calculated. The feature amount in the horizontal direction and the feature amount in the vertical direction are set as projection feature amounts.
The feature amount correlation coefficient calculation unit 225 performs horizontal and vertical directions on the feature amount of the imaging surface white line data calculated by the feature amount projection unit 224 and the feature amount of the video white line data calculated by the white line data imaging surface projection unit 222. To calculate a correlation coefficient indicating the degree of coincidence of feature amounts.
The correlation coefficient determination unit 226 (an example of the feature information specifying unit) calculates imaging surface white line data in which the position of the white line matches the video white line data based on the correlation coefficient calculated by the feature amount correlation coefficient calculation unit 225. Identify.
The vehicle position calculation unit 227 (an example of a second positioning unit) is configured to display the three-dimensional coordinates of the white line indicated by the white line data before projection of the imaging surface white line data specified by the correlation coefficient determination unit 226 and the pixel position of the white line in the camera image. Based on the relationship, the detailed position (positioning position B) of the vehicle 100 is calculated.

FIG. 4 is a flowchart showing a vehicle position specifying method of vehicle position specifying device 200 in the first embodiment.
A vehicle position specifying method executed by the vehicle position specifying device 200 in the first embodiment will be described below with reference to FIG.
Each part of the vehicle position specifying device 200 executes processing described below using a CPU.

<S110: Image processing, imaging processing>
First, the video white line extraction unit 223 extracts a white line portion from the camera video.
At this time, the video white line extraction unit 223 inputs the camera video captured when the GPS receiver 120 receives the positioning signal from the TV camera 110, detects a white line reflected in the input camera video, Video white line data indicating a two-dimensional position of the white line (for example, a pixel position or a two-dimensional coordinate on the imaging surface) is generated. For example, the video white line extraction unit 223 extracts white line edge portions and feature points based on the binarized camera video and the color information of each pixel of the camera video, and specifies the position in the camera image.

Here, the relationship between the camera image captured by the TV camera 110 and the actual position of the white line on the road surface will be described with reference to FIGS. 5, 6, and 7.
5, 6 and 7 are diagrams showing the relationship between the position of the white line on the road surface captured by the TV camera 110 of the vehicle 100 in Embodiment 1 and the position of the white line on the imaging surface.
In FIG. 5, a TV camera 110 is installed in a vehicle 100 with an installation angle θ (elevation angle θ ₁ , azimuth angle θ ₂ , rotation angle θ ₃ ), and is a road located in the traveling direction of the vehicle 100 (the azimuth angle direction of the vehicle 100). The surface is imaged in the range of the field angle φ (vertical direction φ ₁ , horizontal direction φ ₂ ).
For example, in FIG. 5, the TV camera 110 captures two feature points A and C of a stop line marked on the road surface and a white line feature point B displayed on the road surface at a focal distance F away from the lens center O. Two-dimensional image data of the imaging surface T when projected onto T is obtained as a camera image.
In FIG. 6, the road surfaces imaged at the L1 point, the L2 point, and the L3 point indicate the camera images shown in FIG. 7 according to the camera characteristics and parameters (for example, the focal length F, the angle of view φ, and the installation angle θ), respectively. (L1), camera image (L2), and camera image (L3) are displayed. For example, the feature point B of the white line on the road surface imaged at the point L1 shown in FIG. 6 is displayed as the feature point b in the camera video (L1) shown in FIG. Further, for example, the feature points A and C of the road line on the road surface and the feature point B of the white line on the road surface imaged at the point L2 shown in FIG. 6 are feature points a and c in the camera image (L2) shown in FIG. And b. Further, for example, the feature point A of the road stop line imaged at the point L3 shown in FIG. 6 is displayed as the feature point a in the camera image (L3) shown in FIG. In FIG. 7, a dotted line portion is a road surface located outside the angle of view φ of the TV camera 110, and indicates a road surface of a portion that is not displayed in the camera image.

The camera image is obtained by projecting an actual three-dimensional space onto a two-dimensional imaging surface T according to the characteristics and parameters (hereinafter referred to as characteristics) of the TV camera 110 such as the focal length F, the angle of view φ, and the installation angle θ. It is. For this reason, the actual 3 features on the camera image are backprojected to the three-dimensional space according to the characteristics of the TV camera 110 with respect to the position, length, width and tilt of the feature in the camera image. A relative position with respect to the TV camera 110 in the dimensional space can be calculated. In other words, the relative position of the TV camera 110 with respect to the feature reflected in the camera image can be calculated. That is, the absolute position of the TV camera 110 can be calculated if the absolute position (actual three-dimensional coordinates) of the feature reflected in the camera can be specified.
Therefore, the vehicle position specifying device 200 according to the first embodiment specifies white line data corresponding to the white line appearing in the camera image, and the three-dimensional position and white line data in the camera image are determined for the three feature points forming the white line. Based on the actual three-dimensional coordinate values shown, the absolute position (real three-dimensional coordinates) of the TV camera 110 (vehicle 100) is calculated by the principle of triangulation according to the characteristics of the TV camera 110. Here, the position of the TV camera 110 fixedly installed on the vehicle 100 indicates the position of the vehicle 100.

  In FIG. 4, the description of the vehicle position specifying method will be continued.

<S120: First positioning process>
The pseudorange positioning unit 210 calculates a positioning position A based on the pseudorange.
At this time, the pseudorange positioning unit 210 inputs the pseudorange observed when the TV camera 110 captures the camera image from the GPS receiver 120, performs GPS navigation positioning based on the input pseudorange, and performs pseudorange. A positioning position A which is a positioning result based on is calculated. The GPS receiver 120 receives positioning signals from four or more GPS satellites, and for each positioning signal, the time from when the GPS satellite transmits a positioning signal to when the GPS receiver 120 receives it (positioning signal propagation time) Is multiplied by the speed of light to calculate the pseudorange. The pseudo distance indicates the distance from the GPS satellite with respect to the GPS receiver 120. Therefore, the pseudorange positioning unit 210 uses four pseudoranges to generate four spherical equations with each GPS satellite as the origin and the pseudorange as a radius, and uses the generated four equations to The intersection is calculated as a three-dimensional coordinate of the pseudo-range positioning unit 210. Here, the position of the GPS receiver 120 fixedly installed on the vehicle 100 indicates the position of the vehicle 100.

<S130: Feature information acquisition processing>
Next, the white line data acquisition unit 221 acquires white line data around the positioning position A from the road database 190.
At this time, the white line data acquisition unit 221 inputs a positioning position A, which is a positioning result based on the pseudo distance, from the pseudo distance positioning unit 210, and uses the positioning position A as a base point in a predetermined direction in the traveling direction (azimuth angle direction) of the vehicle 100. And white line data indicating the three-dimensional coordinates of the white line for the white line located in the predetermined range in the lateral direction of the vehicle 100 (the width direction of the vehicle 100) is acquired from the road database 190. For example, the white line data acquisition unit 221 acquires the white line data using a range corresponding to the observation accuracy of the GPS receiver 120 or a lane during travel of the vehicle 100 as a predetermined range. For example, when the positioning accuracy based on the pseudo distance of the GPS receiver 120 is ± several meters, the white line data acquisition unit 221 indicates the white line of the white line that is marked within 10 m before and after the vehicle 100 for the lane in which the vehicle 100 is traveling. Data is acquired from the road database 190. At this time, it is assumed that the vehicle 100 is traveling on a straight portion of the road. Further, the white line data acquisition unit 221 obtains the azimuth angle of the vehicle 100 from a gyro (not shown), the azimuth angle of the road being traveled, and the like.

<S140: Feature Projection Acquisition Processing>
Next, the white line data imaging surface projection unit 222 projects the white line data onto the imaging surface of the TV camera 110.
At this time, the white line data imaging plane projection unit 222 inputs white line data of the white line around the vehicle 100 acquired by the white line data acquisition unit 221 from the white line data acquisition unit 221, and the TV camera 110 around the vehicle 100 (positioning position A). Assuming the imaging position, the white line indicated by the white line data is represented by the imaging plane T of the TV camera 110 based on the characteristics such as the focal length F, the angle of view φ, and the installation angle θ of the TV camera 110 and the assumed imaging position of the TV camera 110. The imaging surface white line data projected (projected) is generated. That is, the white line data imaging surface projection unit 222 displays a camera image (imaging surface white line data) that should be displayed on the imaging surface T when the white line is imaged from the assumed position of the TV camera 110 according to the characteristics of the TV camera 110. Generate based on data.

  Here, the camera image actually captured at the point L2 (see FIG. 6), projected (projected) on the imaging surface T of the TV camera 110, and projected as an image is shown as an imaging surface T (L2) in FIG. Then, it is assumed that the white line data imaging surface projection unit 222 projects the white line data onto the imaging surface T of the TV camera 110 assuming that the L1 point (see FIG. 6) is the imaging point of the TV camera 110. At this time, the imaging surface T on which the imaging surface T on which the white line data imaging surface projection unit 222 projects the white line data and the imaging surface T (L2) indicating the camera image of FIG. 9 are superimposed and displayed is as shown in FIG. . Here, in FIGS. 9 and 10 and FIGS. 11 to 15 to be described later, in order to distinguish between the camera image and the imaging surface on which the white line data imaging surface projection unit 222 projects the white line data, the white line reflected in the camera image is displayed. (Video white line data) is indicated by a dotted line, and a white line indicated by the white line data (imaging surface white line data) is indicated by a solid line.

  In addition, the white line data imaging surface projection unit 222, for example, of each section when the white line data acquisition unit 221 divides a predetermined range for which white line data is acquired in S130 like a grid at predetermined intervals. Imaging surface white line data is generated by projecting white line data onto the imaging surface T, assuming that the coordinates of the vertices (plots, intersections) are the imaging points of the TV camera 110. For example, the white line data acquisition unit 221 sets the L1 point shown in FIG. 6 as the assumed imaging point of the TV camera 110, and sets the L2 point and L3 point shown in FIG. Generated for each hypothetical imaging point.

<S150: Correlation determination processing (feature information identification processing)>
Next, the feature amount projection unit 224, the feature amount correlation coefficient calculation unit 225, and the correlation coefficient determination unit 226 generate the white line and white line data imaging surface projection unit 222 indicated by the video white line data generated by the video white line extraction unit 223. The degree of correlation (similarity and coincidence) with the white line indicated by the imaging plane white line data for each assumed imaging point is determined, and imaging plane white line data having a high degree of correlation with the video white line data is specified.

FIG. 8 is a flowchart showing the flow of correlation determination processing (S150) in the first embodiment.
The correlation determination process (S150) executed by the feature amount projection unit 224, the feature amount correlation coefficient calculation unit 225, and the correlation coefficient determination unit 226 will be described below with reference to FIG.
The feature amount projection unit 224, the feature amount correlation coefficient calculation unit 225, and the correlation coefficient determination unit 226 perform the following processing described with reference to FIG. 8 using the CPU.

<S210, S220>
First, the feature quantity projection unit 224 calculates the feature quantity in the horizontal direction of the white line on the imaging plane T (the width direction orthogonal to the traveling direction of the vehicle 100) for each of the imaging plane white line data and the video white line data. To do. For example, the portion of the stop line drawn in the horizontal direction on the road surface has a large feature amount in the horizontal direction, the portion of the white line drawn in the vertical direction has a small feature amount in the horizontal direction, and the white line (including the stop line) is In the unscratched part, the horizontal feature amount is almost zero.
Here, the feature quantity projection unit 224 selects one imaging plane white line data from the plurality of imaging plane white line data calculated by the white line data imaging plane projection unit 222, and calculates a horizontal feature quantity for the selected imaging plane white line data. To do.
The feature amount projection unit 224 may calculate the feature amount based on the edge of the white line, the bending point, the direction vector, or the like, or may calculate a combination of various feature amounts (such as a linear sum) as the feature amount.

<S230>
Next, the feature quantity correlation coefficient calculating unit 225 uses the horizontal feature quantity calculated by the feature quantity projecting unit 224 and the horizontal similarity between the white line indicated by the imaging surface white line data and the white line indicated by the video white line data. A correlation coefficient indicating the degree (degree of coincidence) is calculated.

Here, the correlation coefficient of the feature amount will be described with reference to FIGS.
FIG. 9 is a diagram illustrating an example of a camera video (video white line data) in the first embodiment.
10, FIG. 11 and FIG. 12 show an example of the imaging surface T on which the camera image (image white line data) and the white line data (imaging surface white line data) in Embodiment 1 are superimposed and the camera image (image white line data). ) And white line data (imaging surface white line data).
In the following description, it is assumed that the TV camera 110 has taken an image at the point L2 shown in FIG. The video white line data generated by the video white line extraction unit 223 extracting the white line from the camera video imaged at the point L2 is assumed to indicate the imaging surface T (L2) shown in FIG.
Feature points A, B, and C of the white line (stop line) imaged at the point L2 shown in FIG. 6 correspond to the feature points a, b, and c on the imaging surface T (L2) shown in FIG. Also, the white line feature points indicated by the imaging plane white line data corresponding to the white line feature points A, B, and C are a ′, b ′, and c ′, respectively.
9 to 12 show the white line of the video white line data with a dotted line in order to distinguish the white line indicated by the video white line data generated from the camera video and the white line indicated by the imaging surface white line data generated from the white line data. The white line of the interesting line data is indicated by a solid line. In FIGS. 10 and 12, in order to represent the continuity of the white line indicated by each imaging surface white line data, a part of the white line located outside the range of the imaging surface T indicated by each imaging surface white line data is also indicated by a chain line. Show.

For example, when the feature amount projection unit 224 calculates the feature amount by selecting the imaging surface white line data with the assumed imaging point as a point located in the vicinity of the L1 point illustrated in FIG. 6, the white line and the image indicated by the imaging surface white line data are displayed. The white line indicated by the white line data has a relationship like the imaging surface T shown in FIG. That is, the video white line data shows stop line feature points a and c and white line feature points a and b on the imaging surface T, whereas the imaging surface white line data shows only the white line feature point b ′ on the imaging surface T ( The feature points a ′ and c ′ of the stop line are outside the frame of the imaging surface T).
In this relationship, when the feature quantity correlation coefficient calculation unit 225 calculates the horizontal correlation degree (correlation amount, similarity degree, coincidence degree), the correlation degree is represented by a graph as shown in FIG. The correlation degree graph shown in FIG. 10 shows that when the imaging surface T is scanned in the horizontal direction (lateral direction), the correlation degree of the portion where the feature point b ′ of the imaging surface white line data is low is low, and the feature point of the video white line data is The correlation degree of the part where a and c are located is very low, and the correlation degree of the part where the feature point b of the video white line data is located is high.
The correlation coefficient calculated by the feature amount correlation coefficient calculation unit 225 corresponds to, for example, an integrated value (total value of correlation degrees) of the correlation degree graph shown in FIG.

  Further, for example, when the feature amount projection unit 224 selects imaging plane white line data with a point located in the vicinity of the L2 point illustrated in FIG. 6 as the assumed imaging point, the video white line data and the imaging plane white line data are illustrated in FIG. As shown, the feature points a, b, and c and the feature points a ′, b ′, and c ′ are in a relationship indicating approximately the same position in the horizontal direction. And, as shown in FIG. 11, since the correlation degree of the part where the feature points a, c, a ′ and c ′ are located is very high and the correlation degree of the part where the feature points b and b ′ are located is also high, The correlation coefficient calculated by the quantity correlation coefficient calculation unit 225 is a large value.

  Similarly, when the feature amount projection unit 224 selects imaging plane white line data in which a point located in the vicinity of the L3 point illustrated in FIG. 6 is the assumed imaging point, in FIG. 12, the portion where the feature points a and c are located Since the degree of correlation is very low between the portions where the feature points a ′ and c ′ are located, the correlation coefficient calculated by the feature amount correlation coefficient calculation unit 225 is a small value.

  In FIG. 8, the description of the correlation determination process (S150) is continued.

<S240>
In FIG. 8, the correlation coefficient determination unit 226 compares the correlation coefficient calculated by the feature amount correlation coefficient calculation unit 225 with a predetermined threshold value TH1. That is, the correlation coefficient determination unit 226 compares the similarity (correlation degree, correlation amount) in the horizontal direction (the horizontal direction of the imaging surface T) between the white line indicated by the video white line data and the white line indicated by the imaging surface white line data with a predetermined value. To do.

When the correlation coefficient is equal to or less than TH1 in S240, the feature amount projection unit 224 determines a point located forward or backward in the traveling direction of the vehicle 100 with respect to the assumed imaging point of the imaging plane white line data selected last time as the assumed imaging point. The imaging surface white line data to be selected is selected, and the feature amount in the horizontal direction of the white line indicated by the imaging surface white line data is calculated (S210). For example, the feature amount projection unit 224 is located behind the L2 point in FIG. 6 when the correlation coefficient when selecting the imaging surface white line data with the L2 point shown in FIG. 6 as the assumed imaging point is TH1 or less. A feature amount in the horizontal direction of the white line is calculated for the imaging surface white line data with the L3 point positioned in front of the L1 point and the L2 point as the assumed imaging point.
The feature quantity correlation coefficient calculation unit 225 calculates a correlation coefficient based on the horizontal feature quantity newly calculated by the feature quantity projection unit 224 (S230), and the correlation coefficient determination unit 226 calculates the correlation coefficient. A determination is made (S240).

<S250 to S280>
When the correlation coefficient is larger than TH1 in S240, the feature amount projection unit 224, the feature amount correlation coefficient calculation unit 225, and the correlation coefficient determination unit 226, similarly to the determination of the horizontal direction correlation coefficient in S210 to S240, The feature amount in the vertical direction between the video white line data and the imaging surface white line data is calculated (S250, S260), the correlation coefficient of the calculated feature amount is calculated (S270), and the video white line data and the imaging surface white line data are perpendicular to each other. The direction correlation coefficient is determined (S280), and imaging surface white line data having high correlation in the horizontal direction and the vertical direction with respect to the video white line data is specified.

  FIGS. 13, 14 and 15 show an example of the imaging surface T on which the camera image (image white line data) and the white line data (imaging surface white line data) according to the first embodiment are superimposed and the camera image (image white line data). ) And white line data (imaging surface white line data).

  When the feature quantity projection unit 224 calculates the feature quantity in the vertical direction in S250 for the imaging plane white line data having a positional relationship as shown on the imaging plane T in FIGS. 13 and 15 with respect to the video white line data, the feature quantity phase relationship The correlation coefficient of the feature amount in the vertical direction calculated by the number calculation unit 225 in S270 is low. In addition, when the feature amount projection unit 224 calculates the feature amount in the vertical direction in S250 for the imaging surface white line data having a positional relationship as shown in the imaging surface T in FIG. The correlation coefficient of the feature amount in the vertical direction calculated by the calculation unit 225 in S270 is high.

In S280, the correlation coefficient determination unit 226 compares the correlation coefficient calculated by the feature amount correlation coefficient calculation unit 225 in S270 with a predetermined threshold TH2. That is, the correlation coefficient determination unit 226 compares the similarity (correlation degree, correlation amount) in the vertical direction (the vertical direction of the imaging surface T) between the white line indicated by the video white line data and the white line indicated by the imaging surface white line data with a predetermined value. To do.
If the correlation coefficient is equal to or smaller than TH2 in S280, the feature amount projection unit 224 assumes the imaging that is located in the left-right direction indicating the width direction of the vehicle 100 with respect to the assumed imaging point of the imaging surface white line data previously selected in S250. The imaging plane white line data to be used as a point is selected, and the feature amount in the vertical direction of the white line indicated by the imaging plane white line data is calculated (S250). Then, the feature quantity correlation coefficient calculation unit 225 calculates a correlation coefficient based on the vertical feature quantity newly calculated by the feature quantity projection unit 224 (S270), and the correlation coefficient determination unit 226 calculates the correlation coefficient. A determination is made (S280).

  When the correlation coefficient is greater than TH2 in S280, the imaging plane white line data that has obtained a correlation coefficient greater than TH2 becomes imaging plane white line data having high correlation in the horizontal and vertical directions with respect to the video white line data.

  In the correlation determination process (S150) shown in FIG. 8, imaging surface white line data having a correlation degree higher than a predetermined value is specified. In S130, the white line data acquisition unit 221 acquires from the road database 190, and in S140, white line data imaging is performed. The imaging coefficient white line data having the highest degree of correlation may be specified by calculating the correlation coefficient for all the imaging white line data generated by the surface projection unit 222.

  In FIG. 4, the description of the vehicle position specifying method will be continued.

<S160>
Next, the vehicle position calculation unit 227 specifies the position on the camera image of the white line indicated by the white line data having a high degree of correlation with the camera image.
At this time, the vehicle position calculation unit 227 specifies a two-dimensional position in the camera image corresponding to the feature point for three or more feature points of the white line indicated by the imaging plane white line data specified in the correlation determination processing (S150). .
For example, in FIG. 14, the vehicle position calculation unit 227 uses the white line feature point a of the video white line data as the two-dimensional position in the camera video corresponding to the white line feature points a ′, b ′, and c ′ of the imaging plane white line data. , B and c, the pixel position on the imaging surface T and the two-dimensional coordinate value on the imaging surface T are specified.

<S170: Second positioning process>
Next, the vehicle position calculation unit 227 calculates a positioning position B based on the position of the white line on the camera image and the actual coordinates indicated by the white line data.
At this time, the vehicle position calculation unit 227 calculates the road from the white line data used for the generation (S140) of the imaging surface white line data specified in the correlation determination process (S150) for the three points specifying the two-dimensional position in the camera image. Get the three-dimensional coordinate value on the surface. Next, the vehicle position calculation unit 227 assigns the acquired three-dimensional coordinate value to the two-dimensional position in the camera image specified in S160. Then, the vehicle position calculation unit 227 determines the three-dimensional of the TV camera 110 based on the characteristics of the TV camera 110 such as the focal length F, the angle of view φ, and the installation angle θ and the three-dimensional coordinate values of the three points in the camera video. The coordinates, that is, the three-dimensional coordinates of the vehicle 100 are calculated as the positioning position B. As described above, the self-position can be calculated by triangulation based on the characteristics of the TV camera 110 and the three-dimensional coordinate values of the three points in the camera video.

<S180: Positioning result output process>
Then, the positioning result output unit 230 outputs the positioning position B calculated by the vehicle position calculation unit 227 to a car navigation system or an automatic driving control device.
Thereby, based on the positioning position B with higher accuracy than the positioning position A based on the pseudo distance, the car navigation system can inform the driver of the current position and the route to the destination with high accuracy. Moreover, by obtaining the positioning position B with high accuracy, it is possible to realize control that automatically drives to the destination based on the self position and the road data.

  In the first embodiment, the positioning position A, which is an approximate positioning result for acquiring white line data of the white line located around the vehicle 100, is calculated based on the pseudorange, but a gyroscope or a vehicle speed pulse is used. The positioning position A may be calculated by inertial navigation or dead reckoning. Furthermore, when the GPS receiver 120 can perform GPS observation, the positioning position A is calculated based on the pseudorange, and the GPS receiver 120 can perform GPS observation because the vehicle 100 is traveling in the tunnel. If not, the positioning position A may be calculated by inertial navigation or dead reckoning. Alternatively, the positioning position A may be calculated by a navigation method based on pseudorange, inertial navigation, or a positioning method other than dead reckoning.

Embodiment 2. FIG.
In the second embodiment, the positioning position C obtained by the positioning method obtained by the positioning position A and the positioning method different from the positioning method obtained by the positioning position B are compared with the positioning position B and output as a positioning result of the vehicle 100. A mode in which the reliability of the position is determined based on the comparison result, and the reliability of the determined position of the vehicle 100 based on at least one of the positioning position B and the positioning position C is output as the positioning result. Will be described.
Matters different from those in the first embodiment will be described, and items that will not be described are the same as those in the first embodiment.

FIG. 16 is a functional configuration diagram of the vehicle position specifying device 200 according to the second embodiment.
A functional configuration of the vehicle position specifying device 200 according to Embodiment 2 will be described below with reference to FIG.

  The vehicle position specifying device 200 in the second embodiment is characterized in that the vehicle position specifying device 200 in the first embodiment includes a carrier wave phase positioning unit 240.

  Carrier wave phase positioning section 240 (an example of a third positioning section) calculates the detailed position (positioning position C) of vehicle 100 based on the phase information of the carrier wave observed by GPS receiver 120. For example, the carrier phase positioning unit 240 performs high-precision positioning based on the carrier phase by a method called RTK-GPS (real-time kinematic GPS) or CDGPS (carrier phase differential GPS). Further, for example, the carrier wave phase positioning unit 240 performs positioning using an interference positioning method other than RTK-GPS.

FIG. 17 is a flowchart showing a vehicle position specifying method of vehicle position specifying apparatus 200 in the second embodiment.
A vehicle position specifying method executed by the vehicle position specifying device 200 in the second embodiment will be described below with reference to FIG.

  The vehicle position specifying method according to the second embodiment is characterized in that the third positioning process (S121) is executed with respect to the vehicle position specifying method according to the first embodiment.

<S121: Third positioning process>
The carrier phase positioning unit 240 calculates the positioning position C based on the carrier phase.
At this time, the carrier phase positioning unit 240 receives, from the GPS receiver 120, the phase information (carrier information) of the carrier that carries the positioning signal observed when the TV camera 110 captures the camera image, and based on the carrier phase. GPS navigation positioning is performed, and a positioning position C, which is a positioning result based on the carrier phase, is calculated. As described in the first embodiment, the pseudo-range positioning unit 210 performs positioning based on the pseudo-range between the GPS satellite 120 and the GPS receiver 120 that are measured by multiplying the propagation time of the positioning signal by the speed of light. The carrier phase positioning unit 240 measures the distance between the GPS satellite and the GPS receiver 120 based on the phase of the carrier wave when reaching the GPS receiver 120 and performs positioning.
The distance between the GPS satellite 120 measured based on the carrier phase and the GPS receiver 120 has higher accuracy than the pseudorange, and the positioning result based on the carrier phase also has higher accuracy than the positioning result based on the pseudorange. However, the positioning method based on the carrier phase takes more time to obtain the positioning result than the positioning method based on the pseudorange. Therefore, in the second embodiment, the third positioning process (S121) based on the carrier phase is executed in parallel with the first positioning process (S120) based on the pseudo distance to the second positioning process (S170) based on the white line data. Thus, a positioning position B, which is a highly accurate positioning result based on the white line data, and a positioning position C, which is a highly accurate positioning result based on the carrier phase, are obtained.

<S180: Positioning result output process>
The positioning result output unit 230 compares the positioning position B and the positioning position C to determine the accuracy of the position of the vehicle 100 to be output, and outputs the position of the vehicle 100 and the reliability of the position of the vehicle 100 as the positioning result. To do.

FIG. 18 is a flowchart showing the flow of the positioning result output process (S180) in the second embodiment.
For example, as shown in FIG. 18, the positioning result output unit 230 compares the positioning position B and the positioning position C (S310), and if the positioning position B and the positioning position C match (the position indicated by the positioning position B). And the position indicated by the positioning position C are within a predetermined range), the positioning position C is output as an accurate positioning result, and the positioning position B and the positioning position C do not match (the positioning position B indicates If the point and the point indicated by the positioning position C are not within a predetermined range), the positioning position C is output as an inaccurate positioning result (a positioning result that may not be accurate). In other words, the positioning result output unit 230 outputs the reliability of the positioning position C and the positioning position C, which are accurate or incorrect.
Further, for example, the positioning result output unit 230 may output the positioning position B as the positioning result, may output an intermediate position between the positioning position B and the positioning position C, or based on a predetermined weighting. A position between the positioning position B and the positioning position C may be output. For example, when the positioning position C is weighted more heavily than the positioning position B, the positioning result output unit 230 outputs coordinates indicating a point closer to the positioning position C than the positioning position B. Further, for example, the positioning result output unit 230 may output a value corresponding to the difference between the positioning position B and the positioning position C as the reliability of the positioning position. For example, the positioning result output unit 230 sets the case where the distance between the positioning position B and the positioning position C is 0 as 100% reliability, and the reliability decreases as the distance between the positioning position B and the positioning position C increases (minimum value 0). %).

  Accordingly, the car navigation system and the automatic driving control in which the positioning result output unit 230 outputs the self-position (for example, positioning position C) and the reliability of the self-position (for example, accurate or inaccurate) determined as the positioning result. In the device, it is possible to notify the driver whether or not the navigation accuracy is good, or to switch from the automatic operation mode to the manual operation mode by the driver.

Embodiment 3 FIG.
In the third embodiment, a mode in which the reliability of the position of the vehicle 100 output as a positioning result based on the GPS observation result of the GPS receiver 120 is determined will be described.
Matters different from those of the second embodiment will be described, and matters that will not be described are the same as those of the second embodiment.

FIG. 19 is a functional configuration diagram of the vehicle position specifying device 200 according to the third embodiment.
The functional configuration of the vehicle position specifying device 200 according to Embodiment 3 will be described below with reference to FIG.

  The vehicle position specifying device 200 according to the third embodiment is characterized in that the vehicle position specifying device 200 according to the second embodiment includes a GPS positioning reliability determination unit 250.

The GPS positioning reliability determination unit 250 determines the reliability (accuracy) of the positioning position C measured by the carrier wave phase positioning unit 240 based on the GPS observation result of the GPS receiver 120.
Here, the GPS observation result of the GPS receiver 120 indicates the number of GPS satellites from which the GPS receiver 120 has received a positioning signal, DOP (precision reduction rate), continuity of the positioning result (positioning position C), and the like.
For example, the GPS positioning reliability determination unit 250 compares, for each positioning signal received by the GPS receiver 120, the number of GPS satellites that have transmitted the positioning signal with a predetermined value, and the GPS receiver 120 that has received the positioning signal. When the number of satellites is less than the predetermined value, it is determined that the reliability of the positioning position C is low, and when the number of GPS satellites that the GPS receiver 120 has received the positioning signal is equal to or greater than the predetermined value, the reliability of the positioning position C is Judge as high.
Further, for example, the GPS positioning reliability determination unit 250 calculates a rate of accuracy decrease (DOP) according to the geometric positional relationship between the GPS receiver 120 and the GPS satellite based on the satellite orbit parameter indicated by the positioning signal, and DOP It is determined that the reliability of the positioning position C is low when is below a predetermined value, and the reliability of the positioning position C is high when DOP is greater than the predetermined value.
For example, the GPS positioning reliability determination unit 250 compares the previous positioning position C with the current positioning position C, and the difference between the previous positioning position C and the previous positioning position C is a predetermined value (for example, the vehicle 100 indicated by the vehicle speed pulse). If the reliability of the positioning position C is low, and if the difference between the positioning position C between the previous time and the current time is less than a predetermined value, the reliability of the positioning position C is high. judge.

FIG. 20 is a flowchart showing a flow of positioning result output processing (S180) in the third embodiment.
In the positioning result output process (S180), as shown in FIG. 20, the positioning result output unit 230 compares the positioning position B with the positioning position C (S310), and the positioning position B and the positioning position C match. Outputs a positioning position C as an accurate positioning result. Further, when the positioning position B and the positioning position C do not match, the positioning result output unit 230 refers to the determination result of the GPS positioning reliability determination unit 250 (S320), and the determination result of the GPS positioning reliability determination unit 250 Indicates that the reliability of the positioning result C is high, the positioning position C is output as an accurate positioning result, and the determination result of the GPS positioning reliability determination unit 250 indicates that the reliability of the positioning result C is low. Outputs a positioning result B as an inaccurate positioning result.

  As a result, the positioning result output unit 230 can output, as a positioning result, one of the positioning position B and the positioning position C that seems to have higher reliability.

FIG. 3 is a device configuration diagram relating to positioning of the vehicle 100 on which the vehicle position specifying device 200 according to the first embodiment is mounted. FIG. 3 is a diagram illustrating an example of hardware resources of the vehicle position specifying device 200 according to the first embodiment. FIG. 3 is a functional configuration diagram of the vehicle position specifying device 200 according to the first embodiment. 4 is a flowchart showing a vehicle position specifying method of vehicle position specifying device 200 in the first embodiment. The figure which shows the relationship between the position of the white line on the road surface which the TV camera 110 of the vehicle 100 in Embodiment 1 images, and the position of the white line on an imaging surface. The figure which shows the relationship between the position of the white line on the road surface which the TV camera 110 of the vehicle 100 in Embodiment 1 images, and the position of the white line on an imaging surface. The figure which shows the relationship between the position of the white line on the road surface which the TV camera 110 of the vehicle 100 in Embodiment 1 images, and the position of the white line on an imaging surface. 5 is a flowchart showing a flow of correlation determination processing (S150) in the first embodiment. FIG. 4 is a diagram illustrating an example of a camera video (video white line data) in the first embodiment. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. Example of imaging surface T on which camera video (video white line data) and white line data (imaging surface white line data) are superimposed and displayed in Embodiment 1, and the camera video (video white line data) and white line data (imaging surface white line data) FIG. The function block diagram of the vehicle position specific | specification apparatus 200 in Embodiment 2. FIG. 9 is a flowchart showing a vehicle position specifying method of vehicle position specifying device 200 according to the second embodiment. The flowchart which shows the flow of the positioning result output process (S180) in Embodiment 2. FIG. FIG. 10 is a functional configuration diagram of a vehicle position specifying device 200 according to a third embodiment. The flowchart which shows the flow of the positioning result output process (S180) in Embodiment 3.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 Vehicle, 110 TV camera, 120 GPS receiver, 121 GPS antenna, 190 road database, 200 Vehicle position specifying device, 210 Pseudo distance positioning unit, 220 Image positioning unit, 221 White line data acquisition unit, 222 White line data imaging surface projection unit 223 Video white line extraction unit, 224 Feature amount projection unit, 225 Feature amount correlation coefficient calculation unit, 226 Correlation coefficient determination unit, 227 Vehicle position calculation unit, 230 Positioning result output unit, 240 Carrier phase positioning unit, 250 GPS positioning Reliability determination unit, 911 CPU, 912 bus, 913 ROM, 914 RAM, 915 communication board, 920 magnetic disk device, 921 OS, 923 program group, 924 file group.

Claims (6)

A first positioning unit for positioning the self-position;
A camera that captures a specific direction when the first positioning unit performs positioning;
An image processing unit that uses a CPU (Central Processing Unit) to identify a specific imaging feature that is reflected in the image in a two-dimensional image of the imaging surface captured by the camera;
A feature database for storing feature information indicating a three-dimensional position of the feature and a three-dimensional shape of the feature;
A feature information acquisition unit for acquiring feature information of each feature located around the self-position based on the self-position measured by the first positioning unit from the feature database;
The feature indicated by the feature information based on the position of the feature indicated by the feature information acquired by the feature information acquisition unit and the shape of the feature is used as a projected feature on the imaging surface of the camera using the CPU. A feature projection unit to project;
Each projected feature projected by the feature projection unit and the specific feature identified by the image processing unit are compared on the imaging plane using a CPU, and the projected feature corresponding to the specific imaging feature A feature information identification unit for identifying
CPU based on the two-dimensional position of the projected feature specified by the feature information specifying unit on the imaging surface and the three-dimensional position of the feature indicated by the feature information of the projected feature specified by the feature information specifying unit A second positioning unit for positioning the self-position using
And a positioning result output unit that outputs the self-position measured by the second positioning unit to an output device as a positioning result. The location device further includes:
A third positioning unit for positioning the self-position by a positioning method different from the positioning method of the first positioning unit and the positioning method of the second positioning unit;
The positioning result output unit
Comparing the self-position B measured by the second positioning unit and the self-position C measured by the third positioning unit, and determining the reliability of the self-position output as a positioning result based on the comparison result; 2. The position specifying apparatus according to claim 1, wherein a reliability of the self position determined as the self position based on at least one of the self position B and the self position C is output as a positioning result. The location device further includes:
A GPS receiver that performs GPS (Global Positioning System) observation,
The first positioning unit roughly measures the self-position based on the pseudorange observed by the GPS receiver,
The third positioning unit determines its own position based on the carrier phase observed by the GPS receiver in parallel with the processing from the positioning process by the first positioning unit to the positioning process by the second positioning unit. 3. The position specifying device according to claim 2, wherein detailed positioning is performed. The positioning result output unit
4. The position specifying device according to claim 3, wherein a reliability of a self-position output as a positioning result is determined based on a GPS observation result of the GPS receiver. The first positioning unit performs the first positioning process for positioning its own position,
The camera performs an imaging process that captures a specific direction when the first positioning unit performs positioning,
An image processing unit performs image processing for specifying a specific imaging feature reflected in the image in a two-dimensional image of the imaging surface when the camera has imaged using a CPU (Central Processing Unit),
The feature information acquisition unit indicates the feature information of each feature located around the self-position based on the self-position measured by the first positioning unit, and indicates the three-dimensional position of the feature and the three-dimensional shape of the feature. Performs feature information acquisition processing acquired from the feature database that stores the object information,
The feature projection unit uses the CPU on the imaging surface of the camera to display the feature indicated by the feature information based on the position of the feature indicated by the feature information acquired by the feature information acquisition unit and the shape of the feature. Perform the feature projection process to project as projected features,
The feature information specifying unit compares each projected feature projected by the feature projecting unit with the specific feature specified by the image processing unit using the CPU on the imaging surface, and the specific imaging feature. Perform the feature information identification process to identify the projected feature corresponding to
The two-dimensional position of the projected feature specified by the feature information specifying unit by the second positioning unit and the three-dimensional feature indicated by the feature information of the projected feature specified by the feature information specifying unit Based on the position, the CPU performs a second positioning process for positioning its own position,
A positioning result output process in which a positioning result output unit outputs a self-position measured by the second positioning unit as a positioning result to an output device.   A position specifying program for causing a computer to execute the position specifying method according to claim 5.

Images