JP2019066631A - Map data structure - Google Patents

Abstract

The present invention provides a map data structure that enables height direction recognition in a two-dimensional map.
A map data structure of a map used for traveling a vehicle, including a plurality of two-dimensional map data divided into areas, in which an area where a first road and a second road overlap in a height direction , And the first road and the second road are represented by different two-dimensional map data.
[Selected figure] Figure 7

Description

  The present invention relates to map data structures.

  BACKGROUND In recent years, vehicles (autonomously driven vehicles) that travel autonomously while acquiring the situation around the subject vehicle by various sensors such as a camera mounted on the vehicle and a light detection and ranging (LiDAR) have been developed.

  In order for this type of autonomous driving vehicle to travel by autonomous driving, map data having accurate road information is required (see, for example, Patent Document 1). Accurate road information includes, for example, information such as road inclination, road unevenness, road shoulder unevenness, as well as detailed position information such as road width, lanes, signs, etc., and is generally three-dimensional. It is generated as map data.

JP, 2016-43747, A

  The above-described three-dimensional map for automatic driving is expensive to create. In addition, since such a three-dimensional map for automatic driving is generated using, for example, point cloud information and the like acquired by a rider or the like, the amount of data increases and a large-capacity storage device or the like is required. Therefore, it has been considered to perform automatic driving using a two-dimensional map based on point cloud information and the like.

  When automatic driving is performed using a two-dimensional map, this two-dimensional map is raster information generated by projecting point group information, and thus there is no information in the height direction. Therefore, for example, when traveling at a position where roads having different heights such as overpass and underpass overlap, there is a possibility that the automatically driven vehicle can not correctly recognize the position.

  The problem to be solved by the present invention is, for example, to identify roads overlapping in the height direction in the two-dimensional map as described above.

  In order to solve the above problems, the invention according to claim 1 is a map data structure of a map used for traveling a vehicle, including a plurality of two-dimensional map data divided into areas, and a first road In a region where the second road and the second road overlap in the height direction, the first road and the second road are represented by different two-dimensional map data.

  The invention according to claim 6 is a map data generation method that is executed by a map data generation device that generates map data, and acquires a traveling locus along which a vehicle that obtains recognition information that is a recognition result of a feature traveled. A traveling locus acquisition step, an associating step for associating the traveling locus with a section included in the road network information, a recognition information acquiring step for acquiring the recognition information for each section associated in the associating step, and the recognition information acquisition And generating a two-dimensional map data based on the recognition information acquired in the step.

It is a schematic block diagram of a system which has a map data storage device concerning a 1st example of the present invention. It is a functional block diagram of the server apparatus shown by FIG. It is a flowchart of the production | generation operation | movement of two-dimensional map data in the server apparatus shown by FIG. It is an explanatory view of association with a link and a run track. It is explanatory drawing of the link which cross | intersects. It is an explanatory view of generation of two-dimensional map data. It is an arrangement | positioning image figure of the different two-dimensional map data of the height direction in the same point. It is explanatory drawing of the link used as the object of the map data generation method concerning the 2nd Example of this invention. It is explanatory drawing of the example of division | segmentation of the link shown in FIG. It is a flowchart of the production | generation operation | movement of two-dimensional map data concerning the 2nd Example of this invention. It is an example of two-dimensional map data generated by executing the flowchart shown in FIG. It is a flowchart until it starts driving | running | working, when an autonomous driving vehicle performs an automatic driving | operation.

  Hereinafter, the map data structure according to the embodiment of the present invention will be described. A map data structure according to an embodiment of the present invention includes a plurality of two-dimensional map data divided into areas, and the first road and the second road are overlapped in an area where the first road and the second road overlap in the height direction. The two roads are represented by different two-dimensional map data. By doing in this way, since the two-dimensional map data is different for the area where the first road and the second road overlap in the height direction, the two-dimensional map data corresponding to each road is It is possible to identify roads overlapping in the height direction by referring.

  In addition, a plurality of two-dimensional map data may be divided into regions for each predetermined section of the road. For example, by having two-dimensional map data for each link, the vehicle travels in a region overlapping with other links. Road (link) can be identified.

  In addition, a plurality of two-dimensional map data may be divided into areas having a certain area, and map data can be configured by arranging areas having a certain area in a tile shape, for example. .

  Also, since each of a plurality of two-dimensional map data has identification information indicating a relative height, for example, in a spiral-shaped road, it is divided into a plurality of sections and each has identification information Then, it becomes possible to correctly recognize the position of oneself.

  Moreover, the map data storage device according to the embodiment of the present invention stores the above-described map data structure. In this way, by storing map data in a server device or the like that distributes map data to an automatically driven vehicle or the like, it becomes possible to identify roads overlapping in the height direction during automatic traveling.

  Moreover, the map data generation method according to the embodiment of the present invention acquires the traveling locus traveled by the vehicle acquiring the recognition information which is the recognition result of the feature in the traveling locus acquiring step and acquires the traveling locus in the associating step. It associates with the section that road network information has. Next, recognition information is acquired for each section associated in the association process in the recognition information acquisition process, and two-dimensional map data is generated based on the recognition information acquired in the recognition information acquisition process in the generation process. By doing this, it is possible to associate a traveling locus when acquiring recognition information required to generate two-dimensional map data such as point cloud information with a link or the like of existing road network information. Therefore, since two-dimensional map data can be generated for each section such as a link, even if the roads overlap in the height direction, individual roads can be identified. In addition, it is also possible to use information such as the elevation of the link. Furthermore, by using the existing road network information, it is possible to reduce the generation cost of map data for automatic driving.

  Also, the generation step may generate two-dimensional map data for each link in the road network information. By doing this, two-dimensional map data is generated for each link, so even if it is a crossing road, if it is different as a link, the road can be identified.

  Further, the method may further include a dividing step of further dividing a section in the road network information into a plurality of sections, and the associating step may perform association for each of the sections divided by the dividing step. In this way, for example, when there is a crossing portion in one link such as a spiral road, the relative height position is correctly recognized by dividing the link into a plurality of parts. Is possible.

  Further, the dividing step may add identification information in the height direction corresponding to each of the divided sections. By doing this, for example, in a spiral road, it is possible to correctly recognize its own position by dividing it into a plurality of sections and having identification information in each of the sections.

  A map data structure and a feature information generation method according to a first embodiment of the present invention will be described with reference to FIGS. As shown in FIG. 1, the server device 1 as the feature data storage device according to the present embodiment can communicate with the information processing device 3 mounted on the measurement vehicle C via the network N such as the Internet. ing.

  The measurement vehicle C is a vehicle that has the information processing device 3 and the sensor 4 and measures (acquires) recognition information such as the surrounding environment by the sensor 4 by traveling a predetermined route, and is a manually driven vehicle It may be any of an autonomous driving vehicle.

  The information processing device 3 uploads information based on the surrounding information detected by the sensor 4 to the server device 1. The information processing device 3 may be configured, for example, as a part of a vehicle-mounted device such as a navigation device, or may be configured as a single device. Alternatively, as long as the sensor 4 can be connected, it may be a portable terminal device such as a notebook PC.

  The sensor 4 is a sensor for recognizing information of the vehicle such as the vehicle position and the surrounding environment (a feature present in the periphery), and includes a camera, a rider, a GPS (Global Positioning System) receiver, and an acceleration. Includes sensors and speed sensors. The camera produces a colored image representing the situation of the outside world. The lidar discretely measures the distance to an object present in the outside world, and recognizes the position, shape, etc. of the object as a three-dimensional point group. The GPS receiver generates latitude and longitude position information that represents the current position of the vehicle. The acceleration sensor detects the acceleration of the vehicle. The speed sensor detects the speed of the vehicle. The sensor 4 may be provided with an inertial measurement unit (IMU: Inertial Measurement Unit), a gyro sensor, or the like for recognizing the posture (direction or the like) of the vehicle and correcting acquired data of another sensor. Furthermore, the sensor 4 may include a raindrop sensor for detecting whether it is raining, a thermometer, and a hygrometer. Information of the vehicle such as the vehicle position detected and acquired by the sensor 4 and information such as the surrounding environment are output to the information processing device 3 as the surrounding information.

  The sensor 4 can obtain a feature recognition result necessary to generate map data. A feature is a concept that includes all natural or artificial objects present on the ground. Examples of features include path features located on the vehicle's path (i.e., the road) and peripheral features located on the periphery of the road. As an example of the route top feature, a road sign, a traffic light, a guardrail, a footbridge, etc. may be mentioned, and the road itself is also included. That is, characters and figures drawn on the road surface, and the shape of the road (road width and curvature) are also included in the route features. In addition, examples of the peripheral features include buildings (houses, stores) and billboards located along the road.

  The functional configuration of the server device 1 as a map data storage device is shown in FIG. The server device 1 includes a control unit 11, a communication unit 12, and a storage unit 13.

  The control unit 11 functions as a CPU (Central Processing Unit) of the server device 1 and controls the entire server device 1. The control unit 11 stores the information uploaded from the information processing device 3 in the storage unit 13 as the acquired information 13a, and generates two-dimensional map data 13c based on the acquired information 13a and the road network information 13b described later. .

  The communication unit 12 functions as a network interface or the like of the server device 1 and receives the information uploaded by the information processing device 3. In the present embodiment, the information processing device 3 uploads the information measured by the sensor 4 by communication, but it is not communication and is transferred to the server device 1 by a storage medium such as a memory card after traveling. It is also good.

  The storage unit 13 functions as a storage device such as a hard disk of the server device 1 and stores the above-described acquired information 13a, road network information 13b, and two-dimensional map data 13c. The road network information 13b includes information on a network represented by a combination of a node which is a node on a traffic network expression and a link connecting the nodes. The road network information 13b may be, for example, map data used in a navigation device for guiding a route to a destination.

  The two-dimensional map data 13 c is map data generated by the control unit 11 based on the acquired information 13 a and the road network information 13 b described later. The two-dimensional map data 13 c in the present embodiment is, for example, raster information generated by projecting point cloud information uploaded from the information processing device 3. In addition, although a present Example demonstrates as a two-dimensional map mainly used for automatic driving, it may be a two-dimensional map for using for not only it but the navigation apparatus of a manual operation vehicle, and various map services.

  In the configurations shown in FIG. 1 and FIG. 2, the server device 1 acquires the acquired information 13a, has the road network information 13b, and generates the two-dimensional map data 13c. It may be distributed to a server device or the like.

  Next, an operation of generating two-dimensional map data (map data generation method) in the server device 1 having the above-described configuration will be described with reference to the flowchart of FIG. The flowchart of FIG. 3 is executed by the control unit 11 of the server device 1. That is, the server device 1 is a map data generation device.

  First, in step S1, the measurement vehicle C is made to travel, and the sensor 4 measures (acquires) information (recognition information as a result of feature recognition and travel locus) based on the peripheral information of the measurement vehicle C. The measurement result is uploaded from the information processing device 3 to the server device 1 and accumulated in the storage unit 13 as acquired information 13a. In the present embodiment, it is assumed that the acquired information 13a includes point cloud information and a traveling locus of the measurement vehicle C. The point cloud information is information of a three-dimensional point cloud recognized by the lidar included in the sensor 4. The travel locus is information generated from position information and the like of the measurement vehicle C by the GPS receiver included in the sensor 4 and includes, for example, latitude and longitude information, altitude information, and time information.

  Next, in step S2, a matching result d1 is generated by matching the traveling track of the measurement vehicle C with a link as a section included in the road network (NW) information 13b. The method of matching will be described with reference to FIG. In FIG. 4, there are three nodes N1 to N3 as the road network information 13b. A link B is between node N1 and node N2, and a link A is between node N2 and node N3. Then, it is assumed that the traveling track (acquisition information 13a) in which the measurement vehicle C travels in the order of the node N1, the link B, the node N2, the link A, and the node N3 is acquired (see dashed line in FIG. 4).

  In this case, the position information of the node N1 and the position information included in the traveling track are collated, and if they match or are within a predetermined range, it is determined that the node N1 has been passed. Thereafter, the position information of the node N2 is compared with the position information included in the traveling track, and when it is in agreement or within a predetermined range, it is determined that the node N2 has been passed. Therefore, it is determined that link B has been traveled from the time included in the travel locus determined to have traveled the node N1 to the time included in the travel locus determined to travel the node N2. Similarly, it is determined that the link A has traveled from the time included in the travel locus determined to have passed through the node N2 to the time included in the travel locus determined to have passed through the node N3. When the link has position information, the link position information may be compared with the traveling track. In this way, the link included in the road network information 13b is associated with the traveling track, and the result of the association is the matching result d1.

  Next, in step S3, two-dimensional map data 13c is generated based on the matching result d1 generated in step S2 and the point cloud information (acquisition information 13a). Two-dimensional map data is generated for each matching (associated) link. For example, in the case of link A, the point cloud information acquired from the time included in the travel locus determined to have passed through the node N2 to the time included in the travel locus determined to have passed through the node N3 is read out Then, the point cloud information is projected by a known method to generate two-dimensional map data. By doing this, it is possible to generate two-dimensional map data with only point cloud information of a portion corresponding to link A. Therefore, for example, as shown in FIG. 5, when there is a link C which can not directly travel between the links that intersect (overlap) the link B, the link B and the link C are different as two-dimensional map data. Since it can be set as dimensional map data (file), the traveling position (height position) can be recognized without mistake at the point where the link B and the link C intersect.

  Also, the two-dimensional map data may be generated by being divided into a plurality of regions in the horizontal direction other than the height direction described above. The division method may be divided for each node, or, as shown in FIG. 6, may be divided into, for example, a rectangular shape (tile shape) having a predetermined length in the latitude and longitude direction. That is, it may be divided into areas having a constant area. The case of FIG. 6 is an example in which FIG. 5 is divided into four regions. In the example of FIG. 6, the link A spans two areas: an area to be the file f1 and an area to be the file f2. The link B spans two areas: an area to be the file f3 and an area to be the file f4. The link C spans two areas: an area to be the file f5 and an area to be the file f6. These areas are generated as files f1 to f6 as different two-dimensional map data.

  For example, the file f1 including the node N2 of the link A and the file f3 including the node N2 of the link B are two-dimensional map data indicating the same area but different files. Also, the file f4 including the node N1 of the link B and the file f5 including the link C are two-dimensional map data indicating the same area but different files.

  FIG. 7 is an image diagram in which a file f4 and a file f5 which are two-dimensional map data indicating the same area are overlapped in the height direction. FIG. 7 shows the case where the link C crosses over the link B as a real road or the like. In this case, the autonomous driving vehicle refers to different files in the case of traveling on the link B and in the case of traveling on the link C. Therefore, even when the link intersects the individual roads It can be identified. That is, in a region where link B (first road) and link C (second road) overlap in the height direction, two-dimensional map data in which link B (first road) and link C (second road) are different. Is represented.

  The two-dimensional map data 13c shown in FIG. 2 has a plurality of files (two-dimensional map data) such as the files f1 to f6 described above, and has a map data structure according to the present embodiment. As the map data structure, two-dimensional map data may simply have a plurality of files, or may be a layer structure including a plurality of files as shown in FIG.

  As apparent from the above description, step S1 is a traveling locus acquisition step, step S2 is an association step, and step S3 is a generation step.

  According to the present embodiment, a file including a plurality of two-dimensional map data (files) divided into areas, for example, in an area in which link B and link C overlap in the height direction, link B and link C are different It is represented by. By doing in this way, the area in which the link B and the link C overlap in the height direction is a different file, so when traveling, by referring to the file corresponding to each road, in the height direction It becomes possible to correctly recognize overlapping roads.

  In addition, the plurality of files may be divided into areas for each predetermined link on the road, and a road (link) traveling can be prevented from being erroneously recognized even in an area overlapping with another link. Also, the plurality of files may be divided into areas having a certain area, and map data can be configured by arranging areas having a certain area in a tile form.

  The server device 1 also stores the two-dimensional map data 13c described above. By storing the map data in order to distribute the map data to the automatically driven vehicle or the like, it is possible to correctly recognize the road overlapping in the height direction at the time of automatic traveling.

  Also, in step S1, the travel locus obtained by the vehicle to obtain recognition information that is the recognition result of the feature travels is acquired, and the travel locus obtained as a result of measurement in step S1 is matched with the link possessed by the road network information 13b. And associate. Next, point cloud information is acquired based on the matching result in step S2 in step S3, and two-dimensional map data 13c is generated based on the point cloud information. By doing this, it is possible to associate the traveling locus when acquiring point cloud information with the existing road network information 13b. Therefore, since the two-dimensional map data 13c can be generated for each link, it is possible to identify individual roads even if the roads overlap in the height direction. In addition, it is also possible to use information such as the elevation of the link. Furthermore, by using the existing road network information, it is possible to reduce the generation cost of map data for automatic driving.

  Also, in step S3, two-dimensional map data 13c is generated for each link in the road network information 13b. By doing this, it becomes possible to generate two-dimensional map data for each link of the road network information 13b.

  Next, a map data structure and a map data generation method according to a second embodiment of the present invention will be described with reference to FIGS. The same parts as those of the first embodiment described above are designated by the same reference numerals and the description thereof will be omitted.

  In the first embodiment, matching processing is performed for each link, but in the case of a spiral link as shown in FIG. 8, for example, crossing points (self-crossings) occur in one link. Therefore, matching processing can not be performed for each link as in the first embodiment. The present embodiment is an embodiment in which the first road and the second road are the same link. That is, the present invention is applicable even if the first road and the second road are the same link.

  Therefore, in the present embodiment, one link is divided into a plurality of sections and subdivided at a point where self-crossing occurs. In the example of FIG. 8, there are two points c1 and c2 at which self-intersection occurs, and the points are divided into four sections as shown in FIG. At this time, the divided section is shown as a ratio to the length of the original link. In the case of FIG. 9, the section shown at the bottom is 0 or more and less than 0.3. This indicates a length from one end of the original link to a position of 30%. Similarly, the second from the bottom is 0.3 or more and less than 0.5 (the length from one end to a position of 30% to 50%), and the third from the bottom is 0.5 or more and 0.8 Less than 0.8, the top is 1 or more. In the example of FIG. 9, when dividing one link into a plurality of sections, the numbers 0 to 1 are used, but the notation method of segmentation is not limited to this. Other numbers, letters, etc. may be used as long as the proportion of the length of each link obtained by dividing one link into a plurality of sections can be known. The position coordinates of the start point and the end point of each divided link may be used, or the length calculated from the position coordinates of the start point and the end point of each divided link may be used.

  A method of generating two-dimensional map data (a method of generating map data) according to the present embodiment will be described with reference to the flowchart of FIG. In the flowchart of FIG. 10, steps S1 and S2, acquired information 13a, and road network information 13b are the same as the flowchart shown in FIG.

  In step S11 after acquiring the travel locus (acquired information 13a), it is determined whether there is a link having a self-crossing as described above, and if there is a link having a self-crossing (if S11 is YES) The process proceeds to step S12, and if there is no link having a self-crossing (if S11 is NO), the process proceeds to step S2. Whether or not it has a self-crossing may be determined based on position information, elevation information, and the like included in the link included in the road network information 13b.

  In step S12, the link is divided into a plurality of sections by the method as shown in FIG. 8 and FIG. The divided sections execute step S2 described in FIG. 3 to perform matching processing for each section. In each section, as shown in FIG. 9, since the ratio to the link length is known, matching with the traveling locus is possible. That is, step S12 is a dividing step of further dividing the link (section) into a plurality of sections.

  Then, when generating two-dimensional map data from the matching result d1 after the matching process (step S3A), as shown in FIG. 11, the identification indicating the relative height to each of the files f10 to f13 for each section Give relative height information as information. In FIG. 11, "0" is given to the file f10, "1" is given to the file f11, "2" is given to the file f12, and "3" is given to the file f13. The relative height information is a continuous integer (numerical value) in FIG. 11, but may be other distinguishable information such as an alphabet. This relative height information makes it possible to easily acquire the next two-dimensional map data when traveling on this link. Also, the relative height can be easily determined. That is, identification information in the height direction is given corresponding to each of the divided sections.

  According to the present embodiment, the plurality of two-dimensional map data (files f10 to f13) each have relative height information indicating height, so for example, in a spiral shaped road, a plurality of sections By dividing and having relative height information for each, it becomes possible to correctly recognize its own position.

  Further, step S12 of dividing the link in the road network information 13b into a plurality of sections is further included, and step S2 associates the sections divided in step S12. In this way, for example, when there is a crossing portion in one link such as a spiral road, the relative height position is correctly recognized by dividing the link into a plurality of parts. Is possible.

  Finally, a usage scene of the two-dimensional map data 13c (map data structure) described above will be described with reference to FIG. FIG. 12 is a flowchart up to the start of traveling when the autonomous driving vehicle performs the autonomous driving. First, route planning is performed by a terminal device or the like mounted on the autonomous driving vehicle (step S21). That is, the start point (for example, the current location) and the end point (destination) are determined. A route search is performed with the determined start point and end point as data d2 (step S22). The route search may be performed by a known route search algorithm. A traveling route is determined from the route candidate d3 extracted as a result of the route search (step S23). The determination method may be selected by the user or the like, or may be automatically determined based on traffic conditions such as traffic congestion.

  Next, necessary two-dimensional map data is acquired from the two-dimensional map data 13c based on the determined route d4 (step S24). The two-dimensional map data 13 c may be not the data acquired from the storage unit 13 of the server device 1, but may be two-dimensional map data stored in a terminal device or the like mounted on the autonomous driving vehicle. Then, traveling is started using the acquired two-dimensional map data d5 (step S25).

  In addition, although the Example mentioned above demonstrated by the position where a road cross | intersects, if it is the state which a 1st road and a 2nd road overlap, it will not restrict to it. For example, even if there is a highway on a general road, it is possible to correctly recognize the road on which the vehicle is traveling by applying the present invention.

  Further, the present invention is not limited to the above embodiment. That is, those skilled in the art can carry out various modifications without departing from the gist of the present invention in accordance with conventionally known findings. Of course, as long as the configuration of the map data structure and the map data generation method of the present invention is provided even by this modification, it is included in the scope of the present invention.

1 Server device (map data storage device, map data generation device)
13c Two-dimensional map data (map data structure)
S1 measurement (traveling track acquisition process)
Matching to S2 link (related process)
S3 Map data generation (recognition information acquisition process, generation process)
S12 division (division process)

Claims (9)

A map data structure of a map used to drive a vehicle,
Contains multiple 2D map data divided into areas,
A map data structure characterized in that the first road and the second road are represented by different two-dimensional map data in a region where the first road and the second road overlap in the height direction.   The map data structure according to claim 1, wherein the plurality of two-dimensional map data are divided into predetermined areas on a road.   The map data structure according to claim 1, wherein the plurality of two-dimensional map data are divided into areas having a certain area.   The map data structure according to any one of claims 1 to 3, wherein each of the plurality of two-dimensional map data has identification information indicating a relative height.   A map data storage device storing the map data structure according to any one of claims 1 to 4. A map data generation method that is executed by a map data generation device that generates map data, comprising:
A traveling locus acquisition step of acquiring a traveling locus traveled by a vehicle that obtains recognition information that is a recognition result of a feature;
An associating step of associating the traveling locus with a section included in road network information;
A recognition information acquisition step of acquiring the recognition information for each of the sections associated in the associating step;
A generation step of generating two-dimensional map data based on the recognition information acquired in the recognition information acquisition step;
A map data generation method comprising:   The map data generation method according to claim 6, wherein the generation step generates the two-dimensional map data for each link in the road network information. The method further includes a dividing step of further dividing the section into a plurality of sections,
The map data generation method according to claim 6, wherein the associating step associates the segments divided by the dividing step.   9. The map data generation method according to claim 8, wherein in the generation step, identification information in a height direction is given corresponding to each of the divided sections.

Images