JP2010152139A - Map information creation device, map information creation method, instrument for measuring position of moving body, and method of measuring position of moving body - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To successively create map information while passing points necessary for a road by mounting a map information creation device 1 on a vehicle 12. <P>SOLUTION: The map information creation device includes: a position detection part 4 for detecting the position of a moving body to create position information; a scene imaging part 5 for creating a plurality of mutually different scene images for imaging scenes overlooking a moving direction of the moving body from different positions; an imaging timing generation part 6 for generating imaging start signals starting imaging of the scene imaging part 5; a map information creation part 7 for storing scene information to a storing medium by creating the scene information relating to position information detected based on the scene images. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a map information creating apparatus that creates map information used in a navigation system, a method for creating the map information, and a mobile body position that measures the position of the mobile body using the map information created by the map information creating apparatus. The present invention relates to a measuring apparatus and a measuring method thereof.

  In recent years, navigation systems mounted on vehicles and the like have become widespread. This navigation system acquires position information from a GPS (Global Positioning System) satellite, etc., reads map data stored in a CD-ROM or DVD based on this position information, and displays the current vehicle position on a display device. To do. Thereby, the user has the convenience that can recognize the position of the own vehicle at a glance.

  Position information acquired from generally available GPS satellites includes an error of 10 m or more. Therefore, when the position information from the GPS satellite is directly associated with the map data, an image in which the vehicle does not travel on the road is displayed. In order to correct this, a map matching process is performed. The map matching process is a process for forcibly correcting the position obtained from the GPS satellite on the road of the map. Map data on the road where the moving object is most likely to be located based on the position information obtained from the GPS satellites and the movement information obtained from the traveling distance and angular velocity obtained from the traveling sensor and gyro sensor mounted on the moving object. This is corrected by forcibly moving the moving body to the center of the selected road. Thereby, it can display so that a mobile body may not remove from a road.

However, in the map matching process, since the position of the moving body is corrected to the center of the road, the position on the road where the moving body is traveling and the lane cannot be specified. Therefore, in order to match the position of the moving body to the actual position, the road surface in the moving direction of the moving body is imaged, the white line drawn on the captured road surface is recognized, and the position accuracy of the moving body is improved by this white line information Has been proposed (see, for example, Patent Document 1).
JP 2007-220013 A

However, when the position of a moving object is detected by the white line recognition method, the white line on the road surface is interrupted, the shadow of the building is generated on the road surface under sunlight, or in the case of rain, etc. May be difficult. Even when white line recognition is possible, if there are a plurality of lanes in the traveling direction, it is often impossible to recognize which lane the vehicle is traveling.

  Also, RTK-GPS (Real Time Kinetic GPS) is known as a system for measuring the position of a moving object with high accuracy. RTK-GPS can receive carrier waves sent from a plurality of GPS satellites, detect phase differences thereof, and detect position information of latitude and longitude. Since the available carrier wave has a frequency of 1575 MHz, the wavelength is about 19 cm. By detecting the phase difference, the position can be detected with an accuracy of the order of cm or less. However, since the wavelength of the carrier wave transmitted from the satellite is short, the carrier wave cycle cannot be specified even if it is received by the receiver. Therefore, a base station is provided, and a signal for specifying a carrier cycle is transmitted from the base station. The user can simultaneously receive the carrier wave from the satellite and the signal transmitted by the base station, specify the cycle of the carrier wave, and obtain the position from the phase difference. However, a receiver for using this system is expensive and generally cannot be used easily.

  Therefore, an object of the present invention is to provide a positioning system that can measure the actual position of a moving object with high accuracy by a simple method.

  In the present invention, the following means have been taken in order to solve the above problems.

  In the map information creation device of the present invention, a position detection unit that detects the position of a moving body to generate position information, and a plurality of different landscape images obtained by capturing a landscape that desires the moving direction of the moving body from different positions. A landscape imaging unit to generate; an imaging timing generation unit that generates an imaging start signal for starting imaging of the landscape imaging unit; and landscape information that associates the detected position information based on the landscape image; A map information creation unit for storing landscape information in a storage medium.

  In addition, the landscape imaging unit captures a landscape in a direction opposite to the moving direction of the moving body from different positions, and further generates a plurality of different landscape images.

  Further, the map information creation unit extracts a feature from the generated landscape image, and stores the feature associated with the position information as the landscape information in the storage medium.

  Further, the imaging timing generation unit generates an imaging start signal when the moving body passes through an intersection.

  In the map information creation method of the present invention, the position detection unit detects the position and the traveling direction of the moving body and generates position information, and the imaging timing generation unit captures the landscape image mounted on the moving body. An imaging start signal generation step for generating an imaging start signal for starting imaging of the image capturing unit, and a landscape imaging unit that captures a landscape in which the moving direction of the moving body is desired from different positions according to the imaging start signal, and is different from each other A landscape imaging step for generating a plurality of landscape images and a map information creation unit generate landscape information associated with the detected position information based on the generated plurality of landscape images, and store the landscape information And a map information creating step of creating map information stored in the section.

  In the imaging start signal generation step, the imaging timing generation unit compares the current position of the mobile body with a preset landscape information creation point, and the mobile body reaches the landscape information creation point. When it is determined that the imaging start signal is generated, the imaging start signal is generated.

  The preset landscape information creation point is an intersection.

  The landscape imaging step includes a step in which the imaging unit captures a landscape in a direction opposite to the moving direction of the moving body from a different position and generates a plurality of different landscape images.

  In the map information creation step, the map information creation unit extracts a feature from the landscape image, generates landscape information in which the detected position information is associated with the extracted feature, and stores the memory. I remembered it in the department.

  In the moving body position measuring apparatus of the present invention, the imaging unit that is mounted on the moving body and that captures the lane drawn on the road surface in the direction in which the moving body travels and the landscape in the direction to generate a real image, A real image feature extraction unit that extracts a feature portion from a real image, a position information detection unit that detects a position and an azimuth angle of the moving body, and a landscape in a running direction from a plurality of spaced positions at a specified point on the road A map information storage unit for storing landscape information generated by imaging the image, and acquiring the landscape information from the map information storage unit using the detected position and azimuth, and included in the acquired landscape information A moving body position estimating unit that estimates a position of a moving body that has captured the real image, from a landscape characteristic specifying unit that identifies a characteristic part to be detected, a characteristic part that is specified from the characteristic part of the real image and the landscape information, The estimated position of the moving body An image processing unit that generates road map image or scene image based on, and so comprises a.

  The designated point on the road is an intersection on the map.

  In the moving body position measuring method of the present invention, a position detecting step for detecting a current position and an azimuth angle of the moving body, and a map information acquiring step for acquiring map information from a map information storage unit based on the current position and the azimuth. And detecting that the moving body has approached a designated point in the acquired map information, and from the map information storage unit, a landscape in a traveling direction from a plurality of separated positions at the designated point. A landscape information acquisition step for acquiring landscape information generated by imaging, and detecting that the mobile body has reached a point designated by the map information, and capturing a landscape in the moving direction of the mobile body A real image acquisition step for acquiring a real image, a feature portion is extracted from the acquired real image, a feature portion is specified from the landscape information, and specified from the feature portion of the real image and the landscape information And was so and a mobile location estimation step of estimating the position of a moving body by the feature.

  The designated point is an intersection on the map.

  In the map information creation device of the present invention, a position detection unit that detects the position of a moving body and generates position information, and generates a plurality of different landscape images obtained by capturing a landscape in which the moving direction of the moving body is desired from different positions A landscape imaging unit, an imaging timing generation unit that generates an imaging start signal for starting the imaging of the landscape imaging unit, and landscape information that associates position information detected based on the landscape image is generated, and the landscape information is stored in the storage medium. And a map information creation unit for storing the information. Thus, every time a moving body for creating a map passes through a specific point on the road, it captures a landscape in which the moving direction is desired, generates landscape information associated with the position information, and stores this, so it is stored easily. And map information can be created quickly.

  Further, in the mobile body position measuring apparatus of the present invention, an imaging unit that is mounted on the mobile body and captures a lane drawn on the road surface in the direction in which the mobile body travels and a landscape in the direction, and a characteristic part from the actual image are provided. Real image feature extraction unit to be extracted, position information detection unit to detect the position and azimuth of the moving body, and generated by capturing a landscape in the direction of the road from a plurality of separated positions at a specified point on the road A map information storage unit that stores landscape information, a landscape feature specifying unit that acquires landscape information from the map information storage unit using the detected position and azimuth, and identifies a feature included in the acquired landscape information; A moving object position estimation unit that estimates the position of the moving object that captured the actual image from the characteristic part specified from the characteristic part of the actual image and the landscape information, and a road map image based on the estimated position of the moving object or Image processing unit that generates landscape images , It was to prepare for the. This makes it possible to measure the current position with high accuracy by simplifying the configuration of the measuring device, and even when driving on a road with multiple lanes, the lane on which the moving body is currently driving is displayed on the display screen. It is possible to improve driving safety.

  Hereinafter, the present invention will be described in detail with reference to the drawings.

(Map information creation device)
FIG. 1 is an explanatory diagram of a map information creation device 1 according to an embodiment of the present invention. FIG. 1A is a schematic diagram showing a state in which the vehicle 12 as a moving body is mounted with the map information creation device 1, and FIG. 1B shows map information while the vehicle 12 as a moving body moves on the road. It is a schematic diagram showing the state which is producing. FIG. 2 is a diagram for explaining the principle when the landscape information created by the map information creation device 1 is used as map information for position detection.

  As shown in FIG. 1 (a), the map information creation device 1 according to the present invention is mounted on a moving body of a vehicle 12, and travels on a road as shown in FIG. 1 (b). When a predetermined point on the road, such as an intersection K, is reached, a landscape in the traveling direction is imaged by the first camera 5a and the second camera 5b that are separated from each other. At this time, the third camera 5c and the fourth camera 5d can simultaneously capture the landscape opposite to the traveling direction. Then, the position detection unit 4 detects the position of the road when the image is taken. The map information processing unit 2 stores the landscape image captured by the camera as the landscape information associated with the position Ps detected by the position detection unit 4 or the position of the camera with the relative distance and orientation determined by the position Ps. 9 The vehicle 12 continues to travel and repeats this at the next predetermined point.

  The landscape information created by the map information creation device 1 is used as position detection information for detecting the position of the moving object. FIG. 2 is a two-dimensional plan view of the vehicle 12 and the landscape 15 in the moving direction of the vehicle 12 as seen from above. The y direction is the moving direction of the vehicle 12. The first camera 5a and the second camera 5b mounted on the vehicle 12 are separated in the x-axis direction. The landscape 15 has two feature points W1 and W2 in the depth direction. The feature points are, for example, corners on the roof when looking up at a building, the tip of a tower, and the like. The viewing angles differ between when the feature points W1 and W2 are viewed from the point Pa of the first camera 5a and when viewed from the point Pb of the second camera 5b. The viewing angle θa viewed from the point Pa is larger than the viewing angle θb viewed from the point Pb. Therefore, if the imaging conditions of the first camera 5a and the second camera 5b are set to be the same and the landscape 15 is captured, the interval Da between the feature points W1 and W2 on the image captured by the first camera 5a is the second. It becomes longer than the distance Db between the feature points W1 and W2 on the image captured by the camera 5b.

Now, assuming that the distance on the image of the feature points W1 and W2 imaged from the point Px under the same conditions is Dx, the actual distance between the point Pa and the point Pb is Lo, and the distance from the point Pa to the point Px is Lx, Equation (1) holds,
Lx = Lo × (Da−Dx) / (Da−Db) (1)
That is, the landscape 15 viewed from the points Pa and Pb is captured in advance, and a landscape image is stored as map information. The points Pa and Pb that are the positions of the camera can be determined with high accuracy based on the relative distance from the position Ps detected by the position detection unit 4. The moving body 28 whose position is to be measured captures the same landscape 15 and generates a landscape image when passing through a line connecting the point Pa and the point Pb. And the landscape image of the point Pa and the point Pb is acquired from map information, This acquired landscape image and the landscape image which the mobile body imaged are compared. Then, the position of the moving body 28 can be easily calculated from the above equation (1). In addition, the feature point extracted from the landscape image as map information can be memorize | stored as landscape information. Since the landscape is a three-dimensional space, the feature points are distributed in the three-dimensional space. If the feature points distributed in the vertical direction are compared, the position of the moving body 28 in the traveling direction (y direction) can also be measured by this method.

  In the above description, it has been described that the position of the moving body can be calculated from the distance between the feature points. However, the present invention is not limited to this. From the landscape information stored as map information, the position of the feature point when the position of the moving object is assumed is estimated, the feature point that can be seen from this estimated position, and the feature point extracted from the actual image captured by the moving object And the position where the difference becomes the smallest may be measured as the position of the moving body.

  FIG. 3 is a functional block diagram showing the configuration of the map information creation device 1 according to the embodiment of the present invention. The map information creation device 1 includes a position detection unit 4, a landscape imaging unit 5, a map information processing unit 2 including an imaging timing generation unit 6 and a map information creation unit 7, and a storage unit 9 including a storage medium 10. ing.

  The position detector 4 receives a carrier wave sent from a GPS satellite and a reference signal sent from a base station, and generates position information of latitude and longitude. This position information has an accuracy of cm order. Therefore, when installed on the vehicle 12, the position of the detection position Ps can be accurately measured. The gyro sensor 4b measures the azimuth angle from the change in the angular velocity of the vehicle. The travel sensor 4c measures the speed and travel distance of the vehicle 12. When the vehicle 12 travels in a tunnel or a valley of a building, a carrier wave may not be received from a GPS satellite. In this case, position information is acquired from the gyro sensor 4b and the travel sensor 4c, and the position and the azimuth angle of the vehicle 12 are corrected so that the position information can be detected.

  The landscape imaging unit 5 includes a first camera 5a, a second camera 5b, a third camera 5c, and a fourth camera 5d. Each camera can use a CCD image sensor or a CMOS image sensor. The 1st camera 5a and the 2nd camera 5b image the scenery of the front which is the moving direction of the vehicle 12 from a mutually different position. The 3rd camera 5c and the 4th camera 5d image the landscape of the back on the opposite side to a moving direction from a mutually different position. Each camera is installed at a position accurately measured from the detection position Ps of the position detection unit 4. In the present embodiment, four cameras are used, but one camera may be moved to each imaging point and imaged, and a plurality of landscape images may be captured by one camera.

  The control unit 3 controls the overall operation of the map information creation device 1. The control unit 3 includes a CPU 3a that executes calculations, a ROM 3b that stores programs and the like, and a RAM 3c that is used as a work area of the CPU. The CPU 3a reads each application program of the map information processing unit 2 from the ROM 3b to the RAM 3c in addition to the main program, and implements each unit of the map information processing unit 2 by executing these.

  The map information processing unit 2 includes an imaging timing generation unit 6 that generates an imaging start signal of the landscape imaging unit 5 and a map information creation unit 7 that stores landscape information in the storage medium 10. Moreover, in order to determine the timing which produces | generates an imaging start signal, the map information acquisition part 8 for displaying a present position can be provided.

  The imaging timing generation unit 6 detects that the vehicle 12 has reached a predetermined landscape information creation point, generates an imaging start signal, and the landscape imaging unit 5 receives the image to capture the landscape. At the same time, the position detector 4 receives the imaging start signal and detects the position where the vehicle 12 is currently traveling. For example, it is determined in advance that a landscape information creation point is stored in the map information stored in the map information storage unit 11 and the landscape information is formed for each intersection on the map. The imaging timing generation unit 6 can generate a landscape image by imaging the front and rear in the moving direction of the vehicle 12 every time the vehicle 12 detected by the position detection unit 4 passes the intersection. The imaging start signal may be generated manually in addition to the case where it is automatically generated based on the position detected by the position detection unit 4 as described above. In this case, the imaging timing generation unit 6 is a switch pressed by the operator.

  The map information creation unit 7 acquires a landscape image captured by each camera from the landscape imaging unit 5, and acquires position information at the time when each camera images from the position detection unit 4. And the landscape information which linked | related the landscape image and the position of the camera which imaged this is produced | generated, and it memorize | stores in the storage medium 10. FIG. The map information creation unit 7 acquires the latitude, longitude, and azimuth angle from the position detection unit 4 at the time when the first camera 5a captured. The position of the first camera 5a between the position detection unit 4 and the reception position Ps of the RTK-GPS 4a is fixed. Therefore, the position of the first camera 5a can be determined based on the latitude, longitude, and azimuth detected by the position detector 4. The same applies to the second camera 5b, the third camera 5c, and the fourth camera 5d. The map information creation unit 7 stores a landscape image captured by each camera in the storage medium 10 in association with the latitude, longitude, and azimuth of each camera.

  Moreover, you may memorize | store and memorize | store the feature part which extracted the feature point from the landscape image which each camera imaged in relation to positional information. According to this, the amount of landscape information can be significantly reduced. Therefore, it is not necessary to extract feature points from the landscape image stored in the map information storage unit when the position is detected by the moving body position measuring device, so that the time for measuring the position can be shortened. .

  The map information acquisition unit 8 acquires map information stored in the map information storage unit 11 of the storage unit 9. The imaging timing generation unit 6 can generate an imaging start signal based on the acquired map information. Moreover, the imaging timing generation part 6 can memorize | store a landscape information creation point separately from map information.

  The storage unit 9 includes a map information storage unit 11 that stores map information for displaying a map, and a storage medium 10 for storing landscape information created by the map information creation unit 7. You may make it memorize | store landscape information in this map information storage part 11 as a memory | storage part which can write in the map information storage part 11. FIG.

  If the map information creation device 1 is configured in this way, it is possible to acquire landscape information in the traveling direction and in the opposite direction every time it passes through a predetermined landscape information creation point. Therefore, in the case of a crossroad intersection as shown in FIG. 1B, landscape information in four directions can be acquired simply by passing the vehicle 12 twice in the vertical direction and the horizontal direction. When there is a median that blocks the scenery on the side of the oncoming vehicle at the center of the road, it may be passed through each runway four times.

(Map information creation method)
FIG. 4 is a basic flowchart showing the map information creation method according to the embodiment of the present invention. First, the vehicle 12 equipped with the map information creation device 1 is run (step S1). The imaging timing generation unit 6 detects that the vehicle 12 has reached a predetermined landscape information creation point, and generates an imaging start signal for causing a plurality of cameras of the landscape imaging unit 5 to start imaging (step) S2). The landscape imaging unit 5 receives the imaging start signal and images the landscape in the traveling direction (step S3). Further, the position detection unit 4 receives the imaging start signal, and detects the position and azimuth including the current latitude and longitude of the vehicle 12 (step S4). The map information creation unit 7 generates landscape information by associating the landscape image of the captured landscape with the position and azimuth angle at which the landscape is captured (step S5). The landscape image captured by each camera is associated with position information of the camera that captured the landscape image. Moreover, the feature part which extracted the feature point and the feature line segment from the landscape image can be used as landscape information. The map information creation unit 7 stores the generated landscape information in the storage medium 10 (step S6) and creates map information. The control unit 3 displays a determination screen as to whether or not to stop the creation of map information on the display unit (step S7). If not canceled (No in step S7), the control unit 3 returns to step S1, and if canceled (step S7). Yes), the process is stopped.

  The imaging timing generation process will be described with reference to FIG. The position detection unit 4 generates position information on the current latitude, longitude, and azimuth of the vehicle 12 (step S10). The imaging timing generation unit 6 acquires position information and determines whether or not the moving body has reached a predetermined landscape information creation point (step S11). Whether or not the mobile object has reached the landscape information creation point is compared with the position information of the landscape information creation point and the position information of the vehicle currently running, and the vehicle is moved from the landscape information creation point to a nearby region having a specific range. 12 may be detected and an imaging start signal may be generated. When it is determined that the vehicle 12 has reached the landscape information creation point (Yes in step S11), the imaging timing generation unit 6 generates an imaging start signal (step S12) and proceeds to step S3 illustrated in FIG. When it is determined that the vehicle 12 has not reached the landscape information creation point (No in step S11), the imaging timing generation unit 6 returns to the position detection step S10.

  The landscape information generation process will be described with reference to FIG. The landscape imaging unit 5 receives the imaging start signal and generates a landscape image. The map information creation unit 7 extracts a feature portion including a line segment and a feature point from the landscape image (step S13). The map information creation unit 7 creates landscape information in which the position information detected by the position detection unit 4 is associated with the feature part (step S14). Thus, the amount of map information can be greatly reduced by using the feature portion as landscape information.

  FIG. 7 shows an example of the information content of the map information generated in this way. The road information is represented by nodes indicated by x and links connecting the nodes, and road attributes are stored in each node. The node N1 stores the landscape information in addition to the longitude X and latitude Y of the center point and the intersection, the number of branches, the number of lanes, the lane width, and the like as attributes of the road. Specifically, the first landscape image, the azimuth angle θ1 representing the image capturing direction of the landscape as position information associated therewith, the longitude X1, the latitude Y1, the second landscape image of the imaged point, and the second landscape image associated therewith Azimuth angle θ2, longitude X2, latitude Y2, third landscape image, azimuth angle θ3, longitude X3, latitude Y3, fourth landscape image associated therewith, and azimuth angle θ4, longitude X4, latitude associated therewith Y4 is stored.

  In this embodiment, landscape information is stored for each intersection. Therefore, in FIG. 7, landscape information is stored in the node N6 in addition to the node N1. Moreover, you may memorize | store as landscape information not for every intersection but for every node, or for every even-numbered node or every odd-numbered node. Moreover, the data of the characteristic part extracted from the landscape image other than the landscape image can be stored as the landscape information.

(First moving body position detection device)
Next, the moving body position measuring apparatus 20 that measures the position of the moving body using the map information created by the map information creating apparatus 1 will be described.

  FIG. 8 is a functional block diagram showing the configuration of the moving object position measuring apparatus 20 according to the embodiment of the present invention. The moving body position measuring apparatus 20 includes a control unit 21 that controls the entire apparatus, a position information detection unit 23 that detects the position and azimuth angle of the moving body, and a position and azimuth angle detected by the position information detection unit 23. A position information processing unit 22 that processes position information on the basis of the information and generates display image data of a road map image; an imaging unit 24 that captures the moving direction of the moving object; The image storage unit 25 includes a map information storage unit 27 that stores map information and landscape information, and a display unit 26 that displays a road map image and a moving object.

  Hereinafter, each component will be described. The control unit 21 includes a CPU 31, a ROM 32 and a RAM 33. The ROM 32 stores a main program for controlling the basic operation of the moving body position measuring device 20 and various application programs for processing position information. The CPU 31 reads out each program from the ROM 32 to the RAM 33, executes a calculation using the RAM 33 as its work area, and controls the operation of the moving body position measuring device 20. The CPU 31 realizes each component by executing an application program corresponding to each component of the position information processing unit 22. That is, each component of the position information processing unit 22 is configured by software.

  The position information detection unit 23 includes a GPS unit 35, a gyro sensor 36, and a travel sensor 37. The GPS unit 35 receives carrier waves sent from a plurality of GPS satellites, and detects the current position including the current latitude and longitude from the arrival time difference between the carrier waves. The gyro sensor 36 detects the azimuth angle of the moving body from the change in the angular velocity of the moving body. The travel sensor 37 detects the travel distance and travel speed of the moving body. Since the carrier wave sent from the GPS satellite is once per second, the position of the moving body is complemented by the gyro sensor 36 and the traveling sensor 37. Even when it is difficult to receive a carrier wave from GPS hygiene due to a tunnel or a building in an urban area, the position of the moving body can be estimated. In addition, a magnetic sensor can be installed together with the gyro sensor 36, and the azimuth angle of a moving body can be specified from geomagnetism. In addition, the position of the moving body detected by the GPS unit 35 can be corrected by receiving radio waves transmitted from the base station. The position information detected by the position information detection unit 23 and the position information of the azimuth are transmitted to the position information processing unit 22.

  The imaging unit 24 images a landscape and a road surface in the moving direction of the moving body. The imaging unit 24 can be configured by a CCD imaging device or a CMOS imaging device. The real image storage unit 25 temporarily stores image data of a real image captured by the imaging unit 24. The display unit 26 displays the map image and the landscape image generated by the position information processing unit 22. The display unit 26 can be configured by a liquid crystal display device or an EL display.

  The map information storage unit 27 stores road map information including landscape information. The landscape information is formed in association with the position information based on a landscape image obtained by capturing a landscape in the runway direction from a plurality of spaced positions at a designated point on the road. The road map information is stored in a mass storage device such as a CD-ROM or DVD. In the road map information, road attributes such as distinction between national roads, prefectural roads, and general roads, and information on the number of lanes and lane width are stored as road information.

  The position information processing unit 22 is based on the position information detected by the position information detection unit 23, the actual image captured by the imaging unit 24, and the landscape information stored in the map information storage unit 27. Is measured with high accuracy. The position information processing unit 22 stores map information based on at least the actual image feature extraction unit 41 that extracts a feature from the actual image captured by the imaging unit 24 and the position information detected by the position information detection unit 23. A landscape feature specifying unit 43 that acquires landscape information from the unit 27 and specifies a feature included in the landscape information, and a mobile that estimates the position of the mobile from the feature of the real image and the feature included in the landscape information A position estimation unit 45 and an image processing unit 46 that generates a road map image based on the estimated position of the moving body are provided.

  The position information processing unit 22 further includes a feature matching processing unit 44 that performs matching processing between the feature portion extracted from the real image and the feature portion specified from the landscape information, and a moving object from the real image captured by the imaging unit 24. A white line recognition unit 47 that detects an offset distance from the white line, a road information of a road on which the moving body is located from a map information storage unit, a map information acquisition unit 42 that specifies the number of lanes and a lane width, and position information detection Based on the position information detected by the unit 23, a map matching processing unit 48 for matching the position of the moving body to the road on the map can be provided.

  Hereinafter, the function of each component of the position information processing unit 22 will be specifically described. The real image feature extraction unit 41 extracts the feature portion of the real image from the real image data captured by the imaging unit 24. The characteristic part of the real image includes, for example, the length of the line segment included in the real image, its coordinates, and the coordinates of the intersection where the line segment intersects the line segment. When buildings and trees are included in the captured real image, the line segment formed by the outline of the building and the window frame and the intersection of these line segments are calculated from the brightness difference and saturation difference of the pixels that make up the actual image. Can be extracted. The length of the line segment and the interval between the intersections vary depending on the imaging position where the actual image is captured. That is, the feature portion extracted from the actual image includes position information of the imaging position on the road. Note that the feature part is not limited to the line segment or the intersection point included in the actual image. For example, a characteristic arc length, arc radius, or arc center coordinates included in the actual image can be used as the feature part. . This is because a natural scene is included in a real image, and thus it may be possible to more accurately express a feature part by extracting an arc than a straight line.

  Based on the position information detected by the position information detection unit 23, the map information acquisition unit 42 acquires map information, road attribute information, and landscape information around the current position of the mobile object. These pieces of information are repeatedly acquired one after another according to the azimuth angle and speed of the moving body.

  The landscape feature specifying unit 43 specifies the feature portion of the landscape image from the landscape information read from the map information storage unit 27. For example, when a landscape image is read out as landscape information, the length of a characteristic line segment included in the image and its coordinates, and the coordinates of an intersection where the line segment and the line segment intersect are used as the feature portion. In addition to the line segment and the intersection, a characteristic arc length, arc radius, or arc center coordinates can be used as the feature portion. As will be described later, since the feature matching process between the feature part of the landscape image and the feature part of the real image is performed, the conditions for extracting the feature part from the landscape image are the conditions for extracting the feature part from the real image, It is desirable to set the same.

  The map information storage unit 27 stores landscape information added to a specific point along the road, for example, a node representing an intersection. Landscape information is comprised from the landscape image imaged along the runway direction of the road from the several spaced position of the said node vicinity. Moreover, you may comprise landscape information from the data of the characteristic part previously extracted from the landscape image. Since the data amount of the feature portion is much smaller than the data amount of the landscape image, the map information storage portion 27 can be configured at a lower cost if the landscape information is formed by the feature portion. Furthermore, the feature extraction process which extracts a feature part from a landscape image when measuring the position of a moving body can be omitted. Therefore, it is possible to estimate the position of the moving body and the feature matching process in a shorter time, shorten the time from capturing the actual image to displaying the position measurement result, and display the driving state close to reality. it can.

  The moving object position estimation unit 45 estimates the position of the moving object from the actual image feature part extracted from the actual image and the feature part of the landscape image included in the landscape information acquired from the map information storage unit 27. Specifically, the imaging point can be calculated from the feature portion of the actual image using the equation (1) described with reference to FIG. When the moving body passes on the line connecting the points Pa and Pb of the camera that captured the landscape image, or when the moving body passes near the points Pa and Pb of the camera, a real image is captured. A feature point is extracted from this real image, and the point Px of the moving object can be calculated using Equation (1).

  Further, the feature matching processing unit 44 performs feature matching processing between the feature part of the landscape information acquired from the map information storage unit 27 and the feature part of the actual image captured at the point where the landscape information is generated. The difference value between the feature part of the landscape image captured from the point Pa (see FIG. 2) and the feature part of the actual image, and the feature part of the landscape image captured from the point Pb (see FIG. 2) and the actual image A difference value from the feature part is calculated. The moving body position estimation unit 45 can estimate an imaging point Px (see FIG. 2) at which the two difference values are the smallest.

  The white line recognition unit 47 recognizes a white line drawn on the road surface of the actual image captured by the imaging unit 24. This will be described with reference to FIG. FIG. 9 shows a white line detected by edge processing of a real image. The edge of the white line is detected by the difference processing of the real image data. Next, edge lines e1 and e2 of the left white line H1 and edge lines e3 and e4 of the right white line H2 are detected by performing Hough transform and inverse transform. Thereby, the offset distance Dof which is the distance from the camera point Px which is the viewpoint of the imaging unit 24 and the center line C1 of the left white line can be obtained. The moving object position estimating unit 45 calculates the position of the moving object from the feature part of the real image and the feature part of the landscape image as described above, and moves with higher accuracy if the calculated position is corrected by the offset distance Dof. The position of the body can be estimated. The number of lanes and the lane width can be acquired from the map information storage unit 27 as road information.

  The map matching processing unit 48 corrects the latitude and longitude position information obtained from the GPS satellite using the road map acquired from the map information storage unit 27. The positional information obtained from the GPS satellites by ordinary means includes an error of about 10 m or more. Therefore, if the position of the moving object is displayed as it is, the moving object may be displayed at a place where no road exists. Therefore, the position information detected by the GPS unit 35 is complemented by the gyro sensor 36 and the traveling sensor 37, and road candidates on the map are extracted based on the complemented position information, and the most probable road as a traveling road is obtained. Correction for forcibly moving a moving object to a high road. Thereby, it is possible to eliminate the unnaturalness that the moving body travels outside the road on the map.

  The image processing unit 46 performs image processing for generating an image to be displayed on the display unit 26. Based on the current estimated position of the moving body, map information is read from the map information storage unit 27, and display image data for generating a display on the display unit 26 with the position of the moving body superimposed is generated.

(Moving object position measurement method)
FIG. 10 is a flowchart showing the moving body position measuring method according to the embodiment of the present invention. FIG. 11 is a flowchart illustrating the offset distance detection method. FIG. 12 is a schematic diagram showing a moving object displayed on the display unit 26 after position estimation.

  When the moving body position measurement device 20 starts measuring the position of the moving body 28, the imaging unit 24 images the road surface and the landscape in the moving direction of the moving body 28. The real image storage unit 25 acquires and stores image data of a real image captured by the imaging unit 24. The real image feature extraction unit 41 acquires the image data of the real image stored in the real image storage unit 25 (step S21).

  The position information detection unit 23 detects the position and azimuth angle of the moving body 28. The GPS unit 35 receives a carrier wave from a GPS satellite and generates position information including the latitude and longitude of the moving body 28 from the arrival time difference and the like. The gyro sensor 36 detects a change in the angular velocity of the moving body 28 and generates azimuth angle information of the moving body (step S22). The azimuth angle may be received from a GPS satellite and corrected by a gyro sensor. Further, the azimuth angle may be detected by a geomagnetic sensor.

  The map information acquisition unit 42 acquires map information associated with the current position and the current azimuth of the moving body 28 from the map information storage unit 27 based on the position information detected by the position information detection unit 23. The map matching processing unit 48 performs map matching processing based on the map information acquired by the map information acquisition unit 42, and corrects the position of the moving body 28 to the center of the road on the map. The image processing unit 46 generates display data of map information that displays the moving body 28 in which the position of the moving body 28 is corrected at the center of the road, and outputs the display data to the display unit 26.

  The moving body position estimation unit 45 determines whether or not the moving body is approaching the landscape information storage point where the landscape information is stored (step S23). When the mobile body position estimation unit 45 detects that the mobile body 28 has approached the landscape information storage point (Yes in step S23), the imaging unit 24 captures an image of the landscape in the moving direction of the mobile body 28. A signal is given to obtain a real image. The position information detection unit 23 generates position information from the position and azimuth of the imaging point of the moving body 28 in synchronization with the imaging signal.

  The real image feature extraction unit 41 extracts a feature portion from a real image of a captured landscape. The landscape feature specifying unit 43 specifies a feature portion from the landscape information acquired from the map information storage unit 27 (step S24). The characteristic portion can be a characteristic line segment of the real image or an intersection where the line segment and the line segment intersect. When the landscape information includes a landscape image, the landscape feature specifying unit 43 extracts and specifies the feature portion from the acquired landscape image. The feature matching processing unit 44 performs a feature matching process between the feature part of the real image and the feature part specified from the landscape information (step S25). The moving body position estimation unit 45 estimates the position of the spot where the actual image is captured from the processing result obtained by the feature matching process. Furthermore, the white line can be recognized from the actual image captured by the white line recognition unit 47, and the offset distance Dof between the recognized white line and the position of the moving body can be detected and corrected.

  When the moving body position estimation unit 45 does not detect that the moving body 28 has approached the landscape information storage point (No in step S23), the correction amount corrected at the immediately preceding landscape information storage point is maintained. Further, the position information detector 23 detects changes in the current position and azimuth of the moving body 28. Further, the white line recognition unit 47 detects the offset distance Dof from the white line, and estimates the current position where these corrections are made (step S27).

  The detection of the offset distance in steps S22 and S27 will be described with reference to FIG. The white line recognition unit 47 detects the offset distance of the moving body in the lane from the image data of the actual image captured by the imaging unit 24. The offset distance is a distance from the center of the left white line drawn on the road surface to the moving body position. By detecting the offset distance, it is possible to further improve the estimation accuracy of the position of the moving body.

  The white line recognition unit 47 detects the edge of the white line drawn on the road surface from the image data of the actual image (step S30). Edge detection is performed by luminance difference processing of pixels constituting an actual image. Next, an edge image in real space is generated from the detected edge of the white line (step S31). Next, the edge line segment is estimated by performing the Hough transform and the inverse transform in the Hough space (step S32). Next, center line segments C1 and C2 of the respective edge line segments are obtained from the estimated four edge line segments e1, e2, e3, e4 (see FIG. 9) (step S33). Next, an offset distance Dof between the current point Px of the moving body 28 and the center line segment C1 is obtained (step S34). The mobile body position estimation unit 45 sets the mobile body position corrected using the offset distance Dof.

  FIG. 12 shows an example in which a map image is displayed based on the estimated moving body position (step S26). As shown in FIG. 12, the moving body 28 is displayed on the lane where the vehicle is currently located. Since the offset distance Dof is estimated using the white line recognition unit 47, the position of the moving body 28 in the lane can also be displayed with high accuracy. As a result, a natural image without a sense of incongruity can be displayed, and driving safety can be improved.

  In the above embodiment, the feature matching process is performed to estimate the current position of the moving body 28. Instead, from the coordinates of the feature portion extracted from the landscape image and the coordinates of the feature portion extracted from the actual image. For example, the imaging point of the actual image may be calculated by performing direct calculation using Expression (1).

(Second moving body position measuring device)
FIG. 13 is a schematic diagram of a map for explaining a moving body position measuring apparatus 20 and a measuring method according to another embodiment of the present invention. In this embodiment, the map information storage part 27 memorize | stores the characteristic part extracted from landscape information or landscape information for every intersection of a road.

  FIG. 13 shows an intersection K1 and an intersection K2 ahead. The map information storage unit 27 stores a landscape image as landscape information regarding each intersection. The landscape image is a landscape that desires the ± X1 direction and the ± Y1 direction from the four positions Q11, Q12, Q13, and Q14 with respect to the intersection K1. Similarly, regarding the intersection K2, it is a landscape in which the ± X2 direction and the ± Y2 direction are desired from the four positions Q21, Q22, Q23, and Q24. In addition, the coordinate (X1, Y1) and the coordinate (X2, Y2) are set to X1, X2 as the moving direction when the moving body 28 travels from the left side to the right side.

  Specifically, landscape images Wx11 and Wy11 obtained by capturing a landscape in which the + X1 direction and the -Y1 direction are desired from the position Q11 of the intersection K1 are created. Landscape images Wx12 and Wy12 obtained by capturing a landscape that desires the + X1 direction and the + Y1 direction from the position Q12 are created. Similarly, landscape images Wx13 and Wy13 are generated from a landscape that desires the -X1 direction and -Y1 direction from the position Q13, and landscape images Wx14 and Wy14 are generated from a landscape that desires the -X1 direction and the + Y1 direction from the position Q14. These eight created landscape images Wx11, Wy11, Wx12, Wy12, Wx13, Wy13, Wx14, and Wy14 are stored in the map information storage unit 27 in association with the intersection K1. Similarly, at the intersection K2, a landscape that desires the + X2 direction and the -Y2 direction from the position Q21, a landscape that desires the + X2 direction and the + Y2 direction from the position Q22, a landscape that desires the -X2 direction and the -Y2 direction from the position Q23, and a position A landscape image in which a landscape in which the −X2 direction and the + Y2 direction of Q24 are desired is captured and stored in the map information storage unit 27 in association with the intersection K2. The same applies to other intersections.

  The map information acquisition unit 42 of the moving body 28 acquires, from the map information storage unit 27, map information in the vicinity where the moving body 28 travels based on the position information detected by the position information detection unit 23. When the control unit 21 of the moving body 28 detects that the moving body 28 has approached the intersection K2 based on the position and azimuth angle detected by the position information detection unit 23, the imaging unit 24 changes the moving direction of the moving body 28. The real image is captured and stored in the real image storage unit 25. The real image feature extraction unit 41 extracts a feature portion from the captured real image. The position of the moving body 28 is corrected to the center of the road on the map by map matching processing. Further, the white line recognition unit 47 recognizes the white line drawn on the road surface and estimates the offset distance Dof. The moving body position estimation unit 45 estimates the current position of the moving body from the corrected position information. In FIG. 13, the moving body 28 is traveling on a two-lane road on one side. Therefore, when the position where the moving object can be present is set and the lane in which the moving object travels is estimated by the feature matching process, the position where the moving object can exist is the two positions of the two lanes corrected by the offset distance Dof. Become.

  Next, the landscape feature specifying unit 43 selectively acquires the landscape images Wx11 and Wx12 from the position and azimuth angle of the moving body 28 detected by the position information detection unit 23. The landscape feature specifying unit 43 extracts each feature from the image data of the landscape image Wx11 and the landscape image Wx12. As already described, the feature part can be configured from, for example, a line segment or an intersection included in the landscape image. The moving body position estimation unit 45 calculates the current position of the real image from the feature part of the real image, the landscape images Wx11 and Wx12, and the positions Q11 and Q12. When the feature portions W′x11 and W′x12 extracted from these landscape images are stored instead of the landscape images Wx11 and Wx12, the extraction step by the landscape feature specifying unit 43 can be omitted. Further, if the calculated current position is corrected by the offset distance from the white line by the white line recognition unit 47, the current position can be estimated with higher accuracy.

  When the current position is estimated by the feature matching processing unit 44, the following processing is performed. The feature matching processing unit 44 performs a feature matching process on the feature part of the actual image and the feature part extracted from the landscape image for each of the possible positions of the moving object 28 estimated by the moving object position estimating unit 45. And output the result of the feature matching process. The moving body position estimation unit 45 estimates, from the feature matching processing result, the position where the feature of the landscape image is closest to the feature of the real image as the current position of the moving body. Based on this estimation result, the image processing unit 46 displays the moving body position on the map image and displays the map information on the display unit 26.

  Thus, since the landscape viewed from the intersection is stored in the map information storage unit 27 as landscape information, the position of the moving body 28 can be estimated with high accuracy without storing a huge amount of data. In the navigation system of the moving body 28, one of the most necessary information for the driver is the lane information at the intersection and the position information of the own moving body. According to the present embodiment, these pieces of information can be presented with a simple configuration. In addition, if feature data including a landscape feature is stored as landscape information, it is not necessary to perform a process of extracting the feature from the landscape image, so that the position of the moving body 28 can be estimated in a short time. it can.

  Moreover, you may make it memorize | store the landscape information of the spot set at predetermined intervals along the road as the landscape information memorize | stored in the map information storage part 27. FIG. There are many intersections in cities, but there are few intersections in areas other than cities. Therefore, if the landscape information is stored as map information for each predetermined interval of the road, the position of the moving body on the road can be estimated with high accuracy even in an area with few intersections.

  Moreover, in the said embodiment, although the point which imaged the scenery was made into four positions Q11, Q12, Q13, and Q14 of the corner | angular part of the intersection K1, it is not limited to this. The four positions Q11, Q12, Q13, and Q14 may be in the lane of the road or may be points that straddle the lane. In addition, when a median strip is erected at the center of the road and, for example, the X1 direction is imaged from the position Q12, if the landscape on the traveling side where the moving body 28 travels is hidden by the median strip, What is necessary is just to set two points and to image the driving | running | working side scenery from two points. In this case, there are 8 points to be imaged. In short, in the feature matching processing between the feature portion of the real image and the feature portion extracted from the landscape image, it is only necessary that the location where the real image is captured can be estimated. Further, the number of landscape images captured at the intersection is not limited to eight. Six landscape images are sufficient for a three-way intersection, and ten landscape images are required for a five-way intersection.

It is explanatory drawing of the map information creation apparatus which concerns on embodiment of this invention. It is a figure for demonstrating the utilization principle of the map information produced by the map information production apparatus of this invention. It is a functional block diagram showing the map information creation apparatus which concerns on embodiment of this invention. It is a flowchart figure showing the map information creation method which concerns on embodiment of this invention. It is a flowchart figure showing the imaging timing production | generation process which concerns on embodiment of this invention. It is a flowchart figure showing the production | generation process of the landscape information which concerns on embodiment of this invention. An example of the map information produced by the map information production apparatus which concerns on embodiment of this invention is represented. It is a functional block diagram showing the mobile body position measuring apparatus which concerns on embodiment of this invention. A white line detected by edge processing of a real image is represented. It is a flowchart figure showing the mobile body position measuring method which concerns on embodiment of this invention. It is a flowchart figure showing the offset distance detection method which concerns on embodiment. It is an example of a display of the display part 26 after position estimation with the moving body position measuring method which concerns on embodiment of this invention. It is a schematic diagram of the map for demonstrating the mobile body position measuring apparatus and measuring method which concern on embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Map information production apparatus 2 Map information processing part 3 Control part 4 Position detection part 5 Landscape imaging part 6 Imaging timing generation part 7 Map information creation part 8 Map information acquisition part 9 Storage part 10 Storage medium 11 Map information storage part 12 Vehicle 20 Moving body position measuring device 21 Control section 22 Position information processing section 23 Position information detecting section 24 Imaging section 25 Real image storage section 26 Display section 27 Map information storage section 28 Moving body

Claims (13)

A position detector that detects the position of the moving body and generates position information;
A landscape imaging unit that generates a plurality of different landscape images captured from different positions of a landscape in which the moving direction of the moving body is desired;
An imaging timing generation unit that generates an imaging start signal for starting imaging of the landscape imaging unit;
A map information creation device comprising: a landscape information associated with the detected position information based on the landscape image; and a map information creation unit that stores the landscape information in a storage medium.   2. The map information creation according to claim 1, wherein the landscape imaging unit captures a landscape in a direction opposite to a moving direction of the moving body from different positions and further generates a plurality of different landscape images. apparatus.   The said map information preparation part extracts the characteristic part from the produced | generated landscape image, and memorize | stores the said characteristic part which linked | related the said positional information on the said storage medium as said landscape information, The said 1 or 2 characterized by the above-mentioned. The map information creation device described in 1.   The map information creation device according to claim 1, wherein the imaging timing generation unit generates an imaging start signal when the moving body passes through an intersection. A position detecting step in which the position detecting unit detects the position of the moving body and the traveling direction to generate position information;
An imaging start signal generation step in which an imaging timing generation unit generates an imaging start signal for starting imaging of a landscape imaging unit mounted on a moving body;
A landscape imaging step in which a landscape imaging unit captures a landscape in which the moving direction of the moving body is desired from different positions according to the imaging start signal, and generates a plurality of different landscape images;
The map information creation unit generates landscape information associated with the detected position information based on the generated plurality of landscape images, stores the landscape information in a storage unit, and creates map information A map information creation method comprising: a creation step;   In the imaging start signal generation step, the imaging timing generation unit compares the current position of the mobile body with a preset landscape information creation point, and the mobile body has reached the landscape information creation point. The map information creation method according to claim 5, wherein when the determination is made, the imaging start signal is generated.   The map information creation method according to claim 6, wherein the preset landscape information creation point is an intersection.   The landscape imaging step includes a step in which the imaging unit captures a landscape in a direction opposite to the moving direction of the moving body from a different position and generates a plurality of landscape images different from each other. 8. The map information creation method according to any one of items 7.   In the map information creation step, the map information creation unit extracts a feature portion from the landscape image, generates landscape information in which the detected position information is associated with the extracted feature portion, and stores it in the storage unit. It memorize | stores, The map information preparation method of any one of Claims 5-8 characterized by the above-mentioned. An imaging unit that is mounted on a moving body and captures a lane drawn on a road surface in a direction in which the moving body travels and a landscape in the direction to generate a real image;
A real image feature extraction unit for extracting a feature from the real image;
A position information detection unit for detecting the position and azimuth of the moving body;
A map information storage unit for storing landscape information generated by imaging a landscape in the direction of the road from a plurality of spaced positions at a designated point on the road;
A landscape feature specifying unit that acquires the landscape information from the map information storage unit using the detected position and azimuth, and that identifies a feature included in the acquired landscape information;
From the feature part of the real image and the feature part specified from the landscape information, a mobile object position estimation unit that estimates the position of the mobile object that captured the real image;
A moving body position measuring apparatus comprising: an image processing unit that generates a road map image or a landscape image based on the estimated position of the moving body.   The mobile body position measuring apparatus according to claim 10, wherein the designated point of the road is an intersection on a map. A position detecting step for detecting a current position and an azimuth angle of the mobile body;
A map information acquisition step of acquiring map information from a map information storage unit based on the current position and orientation;
Detecting that the moving body has approached a designated point in the acquired map information, and capturing a landscape in the traveling direction from a plurality of spaced positions at the designated point from the map information storage unit. A landscape information acquisition step for acquiring landscape information generated by
An actual image acquisition step of detecting that the moving body has reached a designated point of the map information, capturing a real image by capturing a landscape in the moving direction of the moving body, and
A moving object position that extracts a characteristic part from the acquired real image, specifies a characteristic part from the landscape information, and estimates a position of the moving object by the characteristic part of the real image and the characteristic part specified from the landscape information A moving body position measuring method comprising: an estimating step.   The method according to claim 12, wherein the designated point is an intersection on a map.

Images