Data generation device, data generation method and computer program

Abstract

Provided is a data generation device comprising: a first generation unit that, on the basis of elevation data including the elevation values of a plurality of points, and road map data indicating a road using a plurality of nodes and a plurality of links, and generates road elevation data obtained by adding the elevation values of the points respectively corresponding to the plurality of nodes to the nodes; and a second generation unit that, on the basis of the road elevation data, calculates gradient information between arbitrarily defined two nodes among the plurality of nodes, and generates road gradient data obtained by adding the gradient information to the plurality of nodes or the plurality of links.

Description

[Technical Field]

[0001] The present disclosure relates to a data generation device, a data generation method, and a computer program. This application claims priority to Japanese Application No. 2020-191101, filed on November 17, 2020, and incorporates by reference all of the contents of said Japanese application. [Background technology]

[0002] Route search devices (e.g., in-vehicle navigation devices) that appropriately guide a vehicle from a departure point to a destination point are known. The route search device calculates the optimal route from the departure point to the destination point based on predetermined road map data using predetermined route search logic, and guides the passenger along the route resulting from this calculation by images and audio from a display, speaker, etc. The road map data includes, for example, nodes and links assigned to roads throughout the country.

[0003] Patent Document 1 discloses a technology for creating stereoscopic image data as seen from inside a traveling vehicle based on road map data and elevation map data in order to display roads, buildings, etc. included in a road map in an in-vehicle navigation device without damaging the image of those objects. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 10-187029 Summary of the Invention

[0005] The data generation device of the present disclosure is a data generation device including: a first generation unit that generates road elevation data based on elevation data including elevation values ​​of a plurality of points and road map data that shows roads with a plurality of nodes and a plurality of links, in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes; and a second generation unit that calculates gradient information between any two of the plurality of nodes based on the road elevation data, and generates road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links.

[0006] The data generation method of the present disclosure includes a first generation step of generating road elevation data based on elevation data including elevation values ​​of a plurality of points and road map data showing roads with a plurality of nodes and a plurality of links, in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the respective nodes; and a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links. The data generation method includes:

[0007] The computer program of the present disclosure is a computer program for operating a computer as a data generation device, and includes: a first generation step of generating road elevation data in which the elevation values ​​of a plurality of points corresponding to each of the plurality of nodes are assigned to the node, based on elevation data including elevation values ​​of the plurality of points and road map data in which roads are represented by a plurality of nodes and a plurality of links; and a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram illustrating a functional configuration of a route search system according to an embodiment. [Figure 2]FIG. 2 is a flowchart showing each step of the data generation process according to the embodiment. [Figure 3] FIG. 3 is a flowchart showing an elevation data generation process according to the embodiment. [Figure 4] FIG. 4 is a diagram schematically showing each piece of data according to the embodiment. [Figure 5] FIG. 5 is a graph schematically showing road elevation data according to the embodiment. [Figure 6] FIG. 6 is a flowchart showing the correction process according to the embodiment. [Figure 7] FIG. 7 is a flowchart showing an overpass correction process according to the embodiment. [Figure 8] FIG. 8 is a graph showing how the elevation values ​​of nodes are corrected in the elevation correction step according to the embodiment. [Figure 9A] FIG. 9A is an explanatory diagram of a layover correction process according to the embodiment. [Figure 9B] FIG. 9B is an explanatory diagram of the layover correction process according to the embodiment. [Figure 10A] FIG. 10A is an explanatory diagram of a shadowing correction process according to the embodiment. [Figure 10B] FIG. 10B is an explanatory diagram of the shadowing correction process according to the embodiment. [Figure 11] FIG. 11 is a flowchart showing a tunnel road correction process according to a modified example. [Figure 12] FIG. 12 is a graph illustrating an altitude correction process according to a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0009] [Problem to be solved by the invention] With the development of autonomous driving technology and electric vehicle (EV) technology, there is a growing need for guidance on more optimal routes.

[0010] In view of such problems, an object of the present disclosure is to provide data that can guide users to a more suitable route.

[0011] [Effects of the invention] According to the present disclosure, it is possible to provide data that can guide a user to a more suitable route.

[0012] [Description of the embodiments of the present disclosure] The gist of the present disclosure includes at least the following.

[0013] (1) The data generation device of the present disclosure is a data generation device including: a first generation unit that generates road elevation data based on elevation data including elevation values ​​of a plurality of points and road map data showing roads with a plurality of nodes and a plurality of links, in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes; and a second generation unit that calculates gradient information between any two of the plurality of nodes based on the road elevation data, and generates road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links.

[0014] According to the data generating device of the present disclosure, it is possible to provide data (road gradient data) that can provide guidance on a more suitable route.

[0015] (2) The system may further include a third generation unit that generates the elevation data based on SAR data obtained by a synthetic aperture radar. This configuration allows elevation data for a wider area to be acquired in a shorter cycle. Here, SAR is an acronym for Synthetic Aperture Radar.

[0016] (3) The third generator may generate a plurality of pieces of elevation data based on the SAR data observed at different times and dates, and may use a statistical value of the generated plurality of pieces of elevation data as the true value of the elevation data. This configuration makes it possible to average out the influence of weather and the like, and to obtain more accurate elevation data.

[0017] (4) The system may further include a correction unit that corrects the elevation value assigned to the node, and the correction unit may correct the elevation value of the abnormal node, which is the node that satisfies the correction condition, based on the elevation value of a normal node, which is the node adjacent to the abnormal node and does not satisfy the correction condition, and the second generation unit may generate the road gradient data based on the corrected road elevation data. With this configuration, even if the road elevation data includes an elevation value that is different from the actual elevation value, a more accurate elevation value can be obtained by the correction.

[0018] (5) The abnormal node may include a first abnormal node connected to a target link, which is the link corresponding to an elevated road and has a gradient between the start point and the end point that exceeds a first predetermined value; the normal node may include a first normal node connected to a non-target link that is not the target link and adjacent to the first abnormal node; and the correction unit may correct the elevation value of the first abnormal node based on the elevation value of the first normal node.

[0019] Because the maximum gradient of a road is regulated by law, links whose gradient exceeds a first predetermined value often contain road elevation data that differs from the actual elevation value. The correction unit extracts nodes connected to such links as first abnormal nodes and corrects their elevation values, thereby making it possible to obtain more accurate elevation values.

[0020] (6) The correction unit may determine whether the link corresponds to an elevated road based on the reflection intensity of at least one of a point corresponding to the link and a point located within a predetermined distance range from the point in a reflection intensity image that shows the strength of reflected waves returning from the Earth's surface to the satellite.

[0021] By determining whether the road corresponding to the link is an elevated road based on the reflection intensity, the first abnormal node can be extracted in a shorter time.

[0022] (7) The abnormal node may include a second abnormal node corresponding to a point where the local incident angle in the SAR data is less than a second predetermined value, and the normal node may include a second normal node adjacent to the second abnormal node and corresponding to a point where the local incident angle exceeds the second predetermined value, and the correction unit may correct the elevation value of the second abnormal node based on the elevation value of the second normal node.

[0023] When using SAR data, it is necessary to pay attention to layover. The smaller the local incidence angle, the more likely it is that layover has occurred. The correction unit extracts nodes corresponding to points where the local incidence angle is less than a second predetermined value as second anomalous nodes and corrects their elevation values ​​to obtain more accurate elevation values.

[0024] (8) The abnormal nodes may include a third abnormal node corresponding to a point where the local incident angle in the SAR data exceeds a third predetermined value, and the normal nodes may include a third normal node adjacent to the third abnormal node and corresponding to a point where the local incident angle is less than the third predetermined value, and the correction unit may correct the elevation value of the third abnormal node based on the elevation value of the third normal node.

[0025] When using SAR data, attention must be paid to radar shadows. The larger the local incidence angle, the more likely it is that a radar shadow will occur. The correction unit extracts nodes corresponding to points where the local incidence angle exceeds a third predetermined value as third anomalous nodes, and corrects their elevation values ​​to obtain more accurate elevation values.

[0026] (9) The abnormal node is a second abnormal node corresponding to a point where the local incident angle in the SAR data is less than a second predetermined value at which a layover occurs in the SAR data; and a third abnormal node corresponding to a point where the local incidence angle exceeds a third predetermined value at which a radar shadow appears in the SAR data, the third predetermined value being greater than the second predetermined value. The normal nodes may include: a second normal node adjacent to the second abnormal node and corresponding to a point where the local incidence angle exceeds the second predetermined value and is less than the third predetermined value; and a third normal node adjacent to the third abnormal node and corresponding to a point where the local incidence angle exceeds the second predetermined value and is less than the third predetermined value. The correction unit may correct the elevation value of the second abnormal node based on the elevation value of the second normal node, and may correct the elevation value of the third abnormal node based on the elevation value of the third normal node.

[0027] When using SAR data, attention must be paid to layovers and radar shadows. The smaller the local incidence angle, the more likely it is that a layover has occurred. The correction unit extracts nodes corresponding to points where the local incidence angle is less than a second predetermined value as second anomalous nodes and corrects their elevation values, thereby obtaining more accurate elevation values. Furthermore, the larger the local incidence angle, the more likely it is that a radar shadow has occurred. The correction unit extracts nodes corresponding to points where the local incidence angle exceeds a third predetermined value as third anomalous nodes and corrects their elevation values, thereby obtaining more accurate elevation values.

[0028] (10) The abnormal node may include a fourth abnormal node connected to a tunnel link, which is the link corresponding to a tunnel road, and the normal node may include a fourth normal node connected to a non-tunnel link, which is the link adjacent to the fourth abnormal node and corresponds to a link other than the tunnel road, and the correction unit may correct the elevation value of the fourth abnormal node based on the elevation value of the fourth normal node.

[0029] Since SAR data is acquired based on waves reflected from the ground surface, if the road corresponding to the link is a tunnel road, the road elevation data often contains an elevation value that differs from the actual elevation. The correction unit extracts the node connected to such a link as a fourth abnormal node and corrects its elevation value, thereby obtaining a more accurate elevation value.

[0030] (11) The data generation method of the present disclosure is a data generation method comprising: a first generation step of generating road elevation data based on elevation data including elevation values ​​of a plurality of points and road map data showing roads with a plurality of nodes and a plurality of links, in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes; and a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links.

[0031] According to the data generation method of the present disclosure, it is possible to provide data (road gradient data) that can provide guidance on a more suitable route.

[0032] (12) The computer program of the present disclosure is a computer program for operating a computer as a data generation device, and includes: a first generation step of generating road elevation data in which the elevation values ​​of a plurality of points corresponding to each of the plurality of nodes are assigned to the node based on elevation data including elevation values ​​of the plurality of points and road map data in which roads are represented by a plurality of nodes and a plurality of links; and a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data in which the gradient information is assigned to the plurality of nodes or the plurality of links.

[0033] According to the computer program of the present disclosure, it is possible to provide data (road gradient data) that can provide guidance on a more suitable route.

[0034] [Details of the embodiments of the present disclosure] Hereinafter, details of embodiments of the present disclosure will be described with reference to the drawings.

[0035] In recent years, advances in electric vehicle (EV) technology have created a need for route guidance services for EVs. For example, rather than simply providing route guidance from a starting point to a destination, efforts are being made to alleviate passengers' concerns about running out of battery power on the way to their destination by suggesting the optimal time to charge their EV, taking into account various conditions such as road congestion, slope, and weather.

[0036] In particular, unlike gasoline-powered vehicles, EVs experience a significant decrease in battery power when traveling uphill, while charging can increase battery power when traveling downhill. For this reason, information about the inclination of the road (gradient information) is important for predicting the remaining battery power of an EV from its driving route. Gradient information includes, for example, the gradient of the road.

[0037] Calculating road gradients requires elevation data containing elevation values ​​for each point. Elevation data can be obtained from road maps published by the Geospatial Information Authority of Japan and the National Aeronautics and Space Administration (NASA), as well as road maps sold by private companies. However, due to factors such as the lack of road maps containing elevation data in some areas and the discrepancy between the actual road height and the elevation values ​​on the road maps, it is not possible to calculate suitable gradient information using the above-mentioned elevation data that has been conventionally available.

[0038] For example, in mountainous areas, when an elevated road is built over a valley, the actual road height may not match the elevation value of the elevation data. This is because the elevation value of the elevation data indicates the elevation of the ground surface (i.e., the elevation of the valley bottom), not the elevation of the surface of the elevated road. Also, in mountainous areas, when a tunnel road is built to penetrate a mountain, the elevation value of the elevation data indicates the elevation of the ground surface (i.e., the elevation of the mountain surface), not the elevation of the surface of the tunnel road, so the actual road height may not match the elevation value of the elevation data.

[0039] Furthermore, if the elevation data update cycle is long (for example, updates every 1 to 10 years), it is not possible to obtain the actual elevation value for locations where the road height has changed due to mountain cutting or filling after the elevation value has been updated.

[0040] Therefore, the route search system 100 of the present disclosure generates elevation data based on data (hereinafter referred to as "SAR data") obtained by a synthetic aperture radar (SAR), which is obtained for a wider area than a road map including elevation values ​​and is updated at a shorter cycle than a road map including elevation values, and generates gradient data including gradient information based on the elevation data and road map data.

[0041] SAR data is obtained by emitting radio waves (e.g., microwaves) from a satellite toward the Earth's surface and receiving the radio waves reflected by the Earth's surface with the satellite's sensor. SAR data is made public by various regions and countries, including the Japan Aerospace Exploration Agency (JAXA), the European Space Agency (ESA), and the Canadian Space Agency (CSA). The update cycle for SAR data is shorter than that of road maps made public by the Geospatial Information Authority of Japan, for example, every week to several months.

[0042] <<Overall configuration of the route search system>> 1 is a diagram showing the functional configuration of a route search system 100 according to an embodiment. The route search system 100 includes, as its functional configuration, a communication unit 10, a database 20, a data generating device 30, a route search device 40, a display unit 61, and an input unit 62. The route search system 100 includes, as its hardware configuration, a calculation unit (e.g., a CPU, a GPU, etc.), a main memory unit (e.g., a RAM, etc.), storage (e.g., an HDD, an SSD, etc.), a communication interface functioning as the communication unit 10, a display and speaker functioning as the display unit 61, and a keyboard and mouse functioning as the input unit 62. The route search system 100 is, for example, a general-purpose computer or a general-purpose server.

[0043] The route search system 100 functions as a data generating device 30 and as a route search device 40 by the calculation unit executing a predetermined program. A database 20 is stored in the storage of the route search system 100. Note that the database 20 may not be stored within the route search system 100, but may be stored in external storage on the cloud (storage external to the route search system 100), and the route search system 100 may access the external storage.

[0044] The communication unit 10 acquires SAR data from a satellite database 50 via a communication network NT1 (for example, the Internet). The satellite database 50 is, for example, a database of an organization that publishes SAR data. The SAR data acquired by the communication unit 10 is input to the data generating device 30. Note that while FIG. 1 shows an example in which the SAR data acquired from the satellite database 50 is directly input to the data generating device 30, it is also possible to configure the acquired SAR data to be temporarily stored in a database 20, and for the data generating device 30 to read the SAR data from the database 20.

[0045] The database 20 includes an elevation database 21 including elevation data D1, a road map database 22 including road map data D2, a first road elevation database 23 including road elevation data D3, a second road elevation database 24 including correction data D4, and a road gradient database 25 including road gradient data D5.

[0046] The data generating device 30 includes an elevation data generating unit 31 (referred to as a "third generating unit" in the present disclosure), a road elevation data generating unit 32 (referred to as a "first generating unit" in the present disclosure), a correcting unit 33, and a road gradient data generating unit 34 (referred to as a "second generating unit" in the present disclosure). Each of these units 31 to 34 is a functional unit that is realized by a calculation unit executing a predetermined program. The data generating device 30 generates road gradient data D5 using SAR data and road map data D2 as source data by a data generating method described below.

[0047] The route search device 40 predicts the remaining battery charge at each point on the route traveled by the electric vehicle EV1, based on the search request received from the electric vehicle EV1 and the road gradient data D5.

[0048] <<About the data generation process>> 2 is a flowchart showing each step of the data generation process executed by the data generating device 30. Each step of the data generation process will be described below with reference to FIG. When the data generation process is started in the data generating device 30, an elevation data generating step S11 is first executed. In the elevation data generating step S11, the elevation data generating unit 31 generates elevation data D1 including elevation information for each point based on the SAR data.

[0049] FIG. 3 is a flowchart showing an example of a subroutine of the elevation data generation step S11. When generating elevation data D1 from SAR data, various known methods may be applied in addition to the method described below. In the elevation data generation step S11, first, an interference image (interferogram) is generated based on two sets of SAR data acquired by performing two observations of the same point on the Earth's surface (interferogram generation step S21). In the interference image, phase information is folded into a range from 0 degrees to 360 degrees. Next, noise components such as phase singularities are removed by filtering (filter noise removal step S22). For example, the noise components are replaced by the arithmetic mean of their surrounding values. After that, the folded phase information of the interference image is unwrapped into an absolute phase value (actual distance value) (unwrap processing step S23).

[0050] The absolute value of the phase is offset according to various conditions (phase offset step S24). Finally, the absolute value of the phase after the offset is converted into an elevation value (z) on the earth's surface using Digital Elevation Model (DEM) data (DEM data) (phase-height conversion step S25). This generates elevation data D1 including the elevation value (z) of each point (x, y) on the earth's surface as coordinate information (x, y, z). The elevation data D1 generated by the elevation data generation unit 31 is stored in the elevation database 21. This completes the elevation data generation step S11.

[0051] The elevation data generation unit 31 acquires new SAR data at predetermined intervals (e.g., every few weeks) and generates elevation data D1. That is, the elevation data generation step S11 is executed as needed. As a result, the elevation data D1 in the elevation database 21 is updated at each predetermined interval.

[0052] 1 and 2. Next, the road elevation data generating unit 32 generates road elevation data D3 based on the elevation data D1 and the road map data D2 (road elevation data generating step S12, "first generating step" in this disclosure).

[0053] FIG. 4 is a diagram schematically showing each of the data D1 to D5. (A) in FIG. 4 shows elevation data D1. For convenience, (A) in FIG. 4 shows a contour image in which points with the same elevation value (z) are connected by lines. (B) in FIG. 4 shows road map data D2. (C) in FIG. 4 shows road elevation data D3. (D) in FIG. 4 shows correction data D4. (E) in FIG. 4 shows road gradient data D5.

[0054] The road map data D2 is data relating to road configurations and includes a directed graph having multiple nodes N1 and multiple links L1. The road map data D2 is sequentially transmitted from a data center or the like (not shown) to the route search system 100, and the road map database 22 is sequentially updated.

[0055] The plurality of nodes N1 are set, for example, at intersections of roads. In addition to intersections, the plurality of nodes N1 are also set at predetermined intervals (for example, every 10 m) at points along the roads. Each of the plurality of nodes N1 has planar coordinate information (x, y). The coordinate information (x, y) is, for example, the latitude and longitude of the point where the node N1 is set.

[0056] Link L1 is set to connect adjacent nodes N1. Link L1 represents the actual road alignment and driving direction. Link L1 has directionality to represent the driving direction. For one-way roads, only one-directional link L1 is set, and for bidirectional roads, a pair of links L1 with different directions is set. Link L1 includes information on road type and link cost LC. Road type includes, for example, the type of general road or toll road, the type of elevated road, tunnel road or local road, the design speed of the road, and the road classification stipulated by law. Link cost LC is a numerical representation of the burden placed on a vehicle when traveling along link L1.

[0057] In the road elevation data generation step S12, first, the road elevation data generation unit 32 reads out elevation data D1 and road map data D2 from the elevation database 21 and road map database 22. Next, based on the elevation data D1, the road elevation data generation unit 32 extracts the elevation value (z) of the point corresponding to the coordinate information (x, y) of node N1 in the road map data D2. Then, the extracted elevation value (z) is assigned to the coordinate information (x, y) of node N1 to generate node N2. That is, the coordinate information (x, y, z) of the newly generated node N2 includes the elevation value (z). As a result, road elevation data D3 including multiple nodes N2 and multiple links L1 is generated. Finally, the road elevation data generation unit 32 stores the road elevation data D3 in the first road elevation database 23. This completes the road elevation data generation step S12.

[0058] The road elevation data generating unit 32 acquires new elevation data D1 and road map data D2 at predetermined intervals (for example, every few weeks) and generates road elevation data D3. That is, the road elevation data generating step S12 is executed as needed. As a result, the road elevation data D3 in the first road elevation database 23 is updated at each predetermined interval.

[0059] Fig. 5 is a graph that schematically illustrates road elevation data D3. In the graph of Fig. 5, the horizontal axis represents the horizontal distance from a predetermined point A to a predetermined point B, and the vertical axis represents elevation. In the graph of Fig. 5, black circles represent nodes N2 of road elevation data D3, and lines connecting adjacent black circles represent links L1 of road elevation data D3. As a comparative example, Fig. 5 also illustrates elevation data obtained from a road map of the Geospatial Information Authority of Japan (hereinafter referred to as "comparison data") as a dashed line.

[0060] Also, as a reference example, Figure 5 shows elevation data (hereinafter referred to as "GPS data") acquired by a quasi-zenith satellite system when a vehicle equipped with a GPS (Global Positioning System) tracker is actually driven from point A to point B, and the data is shown as a white circle. GPS data is so-called ground truth data that almost accurately represents the actual height of roads. However, since acquiring GPS data requires that a vehicle actually drive on the roads as described above, it is not realistic to acquire GPS data for all roads.

[0061] Let's take a look at the area indicated by the arrow AR1 in Figure 5 (hereafter referred to as "area AR1"). In area AR1, the elevation values ​​in the comparison data drop in a valley-like pattern. In contrast, the elevation values ​​in the GPS data do not show a valley, and area AR1 is shown as a relatively flat road. In fact, the road from point A to point B is a highway that opened two years before the GPS data was acquired, and area AR1 is a newly opened service area. This highway and service area were built after the comparison data was most recently updated, so the elevation values ​​in the comparison data show the valley before the service area was built.

[0062] In contrast, the elevation values ​​of the road elevation data D3 are almost the same as those of the GPS data. This is because the update cycle of the SAR data is shorter than that of the comparison data, and therefore the elevation values ​​of the ground surface after the service area was developed for the area AR1 can be shown as elevation values. In this way, by generating elevation data D1 using SAR data with a short update cycle and generating road elevation data D3 based on that elevation data D1, it is possible to obtain elevation values ​​that are closer to the actual elevation even if changes in the topography occur.

[0063] Next, let us look at the area indicated by the arrow AR2 in Figure 5 (hereinafter referred to as "area AR2"). In area AR2, the elevation values ​​of the comparison data are valley-like. In contrast, the elevation values ​​of the GPS data are shown as a relatively flat road. Area AR2 is actually an elevated road (e.g., a road bridge) built over a valley. The GPS data correctly obtained a flat elevation value because a vehicle traveled on the elevated road. Because the elevated road was built after the most recent update of the comparison data, the comparison data shows the bottom of the valley as an elevation value. Note that if the elevation values ​​of the comparison data indicate a location on the ground rather than a road, the elevation value of a road that is higher than the ground, such as an elevated road, will not be shown as an elevation value in the comparison data, even if the comparison data is updated after the elevated road is completed. In this case, the comparison data for area AR2 will remain valley-like, even if it is updated after the elevated road is completed.

[0064] The elevation values ​​of the road elevation data D3 are almost the same as those of the comparison data. This is because the width of the elevated road is narrower than the resolution of the SAR data, and the elevation of the valley bottom is obtained as the elevation value in the SAR data.

[0065] The portion indicated by the arrow AR3 in Figure 5 (hereinafter referred to as "area AR3") is actually an elevated road. As with area AR2, the elevation values ​​of area AR3 in the GPS data are correctly shown as a relatively flat road, while the elevation values ​​of the comparison data and road elevation data D3 are shown as valleys.

[0066] As described above, the road elevation data D3 may contain elevation values ​​that differ from the actual elevation due to factors such as the resolution of the SAR data. Therefore, in order to bring the elevation values ​​of the road elevation data D3 closer to the actual elevation values, the correction unit 33 performs the correction step S13 (FIG. 2) described below. In the correction step S13, nodes N2 that contain elevation values ​​that differ from the actual elevation are extracted using various correction conditions described below. The elevation values ​​of the extracted nodes N2 are then corrected based on the elevation values ​​of nodes N2 that do not satisfy the correction conditions (i.e., that contain the actual elevation values).

[0067] 6 is a flowchart showing an example of a subroutine of the correction step S13. In this embodiment, the correction unit 33 executes an overpass correction step S31, a layover correction step S32, and a shadowing correction step S33 in this order in the correction step S13. However, the order of steps S31 to S33 is not limited to this, and any order is acceptable. Furthermore, the correction unit 33 may execute only two or one of these steps S31 to S33.

[0068] 7 is a flowchart showing an example of a subroutine of the overpass correction step S31. In the overpass correction step S31, a node N2 that is actually an overpass but is likely to have an elevation value that is not that of an overpass, such as the bottom of a valley, in the SAR data is extracted based on the gradient of the link L1 and the road type, and the elevation value of the node N2 is corrected.

[0069] In Japan, laws and regulations stipulate the maximum gradient for each road type. For example, for a Type 3 ordinary road (e.g., a national highway) with a design speed of 50 km / h, the maximum gradient is stipulated as a general rule to be 6% or less. Similarly, for a Type 1 ordinary road (e.g., an expressway) with a design speed of 80 km / h, the maximum gradient is stipulated as a general rule to be 4% or less. For example, if the road type of link L1 is an expressway with a design speed of 80 km / h, but the gradient g1 of link L1 exceeds 4%, there is a high possibility that the SAR data has mistakenly acquired the elevation of a valley bottom or the like rather than the road. In the elevated road correction step S31, the gradient g1 is calculated for each link L1, and if the gradient g1 exceeds a predetermined maximum value set for each road type, the elevation value of node N2 is corrected.

[0070] First, the correction unit 33 reads out the road elevation data D3 from the first road elevation database 23. Next, the correction unit 33 calculates the gradient g1 between the start point and the end point of the link L1 in the road elevation data D3 (gradient calculation step S41).

[0071] The gradient g1 is obtained by dividing the vertical distance between two points by the horizontal distance and multiplying the result by 100. The unit of gradient g1 is %. For example, if node N2 located at the start point of link L1 has coordinate information (x1, y1, z1) and node N2 located at the end point of link L1 has coordinate information (x2, y2, z2), the gradient g1 of link L1 is expressed by the following formula (1). Note that gradient g1 may also represent the inclination of the surface with respect to the horizontal plane using an "angle." In this embodiment, gradient g1 is calculated for each link L1 as described above. In other words, if there are 100 links L1, 100 gradients g1 are calculated.

[0072]

number

[0073] Next, the correction unit 33 extracts links L1 whose gradient g1 exceeds a first predetermined value (extraction step S42). If the gradient g1 of link L1 exceeds the first predetermined value, the process proceeds to the operator confirmation step S43 (YES route in S42 in FIG. 7). If the gradient g1 of link L1 is equal to or less than the first predetermined value (NO route in S42 in FIG. 7), the overpass correction step S31 ends.

[0074] Here, the first predetermined value is the maximum gradient allowed for the road gradient, and is set for each road type of link L1. The relationship between the road type and the first predetermined value is stored, for example, as table-format data in the database 20 or in a storage area of ​​the data generating device 30. For example, if the road type is a general national highway with a design speed of 50 km / h, the first predetermined value is 6%. Also, if the road type is an expressway with a design speed of 80 km / h, the first predetermined value is 4%.

[0075] Next, the correction unit 33 displays a satellite photograph of a point including the link L1 whose gradient g1 exceeds the first predetermined value on the display unit 61, and receives input from the operator via the input unit 62 (operator confirmation step S43). If the operator visually checks the satellite photograph and confirms that the elevation of the point including the link L1 needs to be corrected (for example, that there is an elevated road at the point including the link L1), the operator inputs this information via the input unit 62. For example, the operator clicks the "Elevated road present" button displayed on the display unit 61 with the mouse. In this case, the process proceeds to the elevation correction step S44 (the YES route of S43 in FIG. 7).

[0076] Furthermore, if the operator visually checks the satellite photograph and confirms that the point including link L1 does not require elevation correction (for example, there is no elevated road at the point including link L1, and there is a road with a gradient steeper than the specified gradient), the operator inputs this information using the input unit 62. For example, the operator clicks the "No elevated road" button displayed on the display unit 61 with the mouse. In this case, the elevated road correction step S31 ends (the route of NO at S43 in FIG. 7).

[0077] If the road type of the link L1 includes information on whether it is an elevated road, the correction unit 33 may perform an elevated road presence / absence determination step instead of the operator confirmation step S43. In this case, the step of the operator visually inspecting the satellite photograph can be omitted. That is, if the road type of the link L1 whose gradient g1 exceeds the first predetermined value is an elevated road, the process proceeds to the elevation correction step S44. If the road type of the link L1 is other than an elevated road, the elevated road correction step S31 is terminated.

[0078] Next, the correction unit 33 corrects the altitude value of the node N2 (altitude correction step S44). FIG. 8 is a graph showing how the elevation value of node N2 is corrected in the elevation correction step S44. In FIG. 8, node N2 connected to link L1 (hereinafter, this link L1 will be referred to as the "target link") that corresponds to an expressway and has a gradient g1 exceeding a first predetermined value is shown as an "abnormal node N13." Link L1 that corresponds to an expressway is a link L1 for which an operator has input indicating that correction is required, or a link L1 whose road type is an elevated road. FIG. 8 includes an area on the point A side where there are four consecutive abnormal nodes N13, and an area on the point B side where there are five consecutive abnormal nodes N13. These areas correspond to areas AR2 and AR3 in FIG. 5, respectively.

[0079] Node N2, which is adjacent to abnormal node N13 via link L1 and is connected to an asymmetric link of link L1 that is not an asymmetric link, is referred to as a "normal node." In FIG. 8, these nodes are shown as normal nodes N11 and N12. Specifically, an asymmetric link is a link L1 that satisfies at least one of the following conditions: it does not correspond to an expressway, and its gradient g1 is equal to or less than a first predetermined value. Normal node N11 is a node adjacent to abnormal node N13 on the point A side (one side). Normal node N12 is a node adjacent to abnormal node N13 on the point B side (other side).

[0080] The correction unit 33 corrects the elevation value of the abnormal node N13 based on the elevation values ​​of the normal nodes N11 and N12. For example, the correction unit 33 changes the elevation value of the abnormal node N13 to the average value of the elevation value of the normal node N11 and the elevation value of the normal node N12. For example, if the normal node N11 has an elevation value z11 and the normal node N12 has an elevation value z12, the corrected elevation value z13 of the abnormal node N13 will be (z11 + z12) / 2. This completes the elevated road correction process S31.

[0081] Here, the original elevation value z13 of the abnormal node N13 tends to be close to the elevation value z11 of the normal node N11 when the abnormal node N13 is located close to the normal node N11, and tends to be close to the elevation value z12 of the normal node N12 when the abnormal node N13 is located close to the normal node N12. For this reason, when correcting the elevation value of the abnormal node N13, the horizontal distance between the normal nodes N11 and N12 and the abnormal node N13 may be taken into consideration. For example, if the horizontal distance between the normal node N11 and the abnormal node N13 is α and the horizontal distance between the normal node N12 and the abnormal node N13 is β, the corrected elevation value z13 will be z11 + ((z11 - z12) / (α + β)) × α.

[0082] By correcting the elevation value of node N2 in the elevation correction step S44, the elevation values ​​of the corrected road elevation data D3 (i.e., corrected data D4) for areas AR2 and AR3 are approximately the same as the elevation values ​​of the GPS data. In this way, even if the road elevation data D3 includes an area (i.e., an abnormal node) with an elevation value different from the actual value due to the resolution of the SAR data, etc., the area can be extracted based on correction conditions such as the gradient of link L1 and road type, and the elevation value of the area can be corrected based on the elevation value of a nearby node N2 (i.e., a normal node) that does not satisfy the correction conditions, thereby making the elevation value closer to the actual elevation value (GPS data). This makes it possible to obtain a more accurate elevation value.

[0083] Refer to FIG. 6. Next, the correction unit 33 executes a layover correction step S32 and a shadowing correction step S33. Here, when using SAR data, attention must be paid to layover and radar shadow. Layover is a phenomenon in which tall buildings or mountains are determined to be close to the satellite in SAR data and are observed to collapse toward the satellite. Radar shadow is a phenomenon in which radio waves emitted from a satellite are blocked by tall buildings or mountains, preventing the radio waves from reaching the area behind them, making it impossible to obtain information about the shadowed area of ​​the mountain or other object. In areas where layover or radar shadow occurs, the elevation value of node N2 is likely to be inaccurate. For this reason, in this embodiment, the layover correction step S32 and the shadowing correction step S33 extract areas where layover or radar shadow occurs and correct the elevation value of node N2 included in those areas.

[0084] 9A and 9B are explanatory diagrams of the layover correction process S32. The smaller the local incident angle (LIA), the more likely a layover has occurred. Here, the local incident angle is the angle between the normal to the object and the line drawn from the object to the satellite. Therefore, the correction unit 33 determines that a layover has occurred for node N2 where the local incident angle is equal to or less than a second predetermined value (e.g., 0 degrees), and performs correction. FIG. 9A is a map showing the local incident angle of each point in the form of contour lines. In FIG. 9A, areas where the local incident angle is equal to or less than 0 degrees are shaded.

[0085] 9B is a graph showing how the elevation value of node N2 is corrected in the layover correction step S32. In FIG. 9B, node N2 at a point where the local incident angle is 0 degrees or less is shown as "abnormal node N23." FIG. 9B includes an area where three consecutive abnormal nodes N23 occur.

[0086] Furthermore, a node N2 adjacent to the abnormal node N23 via the link L1 and having a local incident angle greater than the second predetermined value (i.e., a node unlikely to have experienced a layover) is referred to as a "normal node." In FIG. 9B, these nodes are shown as normal nodes N21 and N22. The normal node N21 is a node adjacent to the abnormal node N23 on one side. The normal node N22 is a node adjacent to the abnormal node N23 on the other side. In this embodiment, the node N2 whose local incident angle is equal to the second predetermined value is determined to be the abnormal node N23, but this node N2 may also be determined to be a normal node. That is, the correction unit 33 may determine a node N2 whose local incident angle is less than the second predetermined value as an abnormal node, and a node N2 adjacent to an abnormal node and having a local incident angle equal to or greater than the second predetermined value as a normal node.

[0087] Furthermore, node N2, which corresponds to abnormal node N26, which is likely to have a radar shadow (described later), may be excluded from the normal nodes N21 and N22. That is, normal nodes N21 and N22 are adjacent to abnormal node N23 and correspond to points where the local incident angle exceeds a second predetermined value and is less than a third predetermined value (described later).

[0088] The correction unit 33 corrects the elevation value of the abnormal node N23 based on the elevation values ​​of the normal nodes N21 and N22. For example, the correction unit 33 changes the elevation value of the abnormal node N23 to the average value of the elevation value of the normal node N21 and the elevation value of the normal node N22. As in the above-mentioned elevation correction step S44, when correcting the elevation value of the abnormal node N23, the horizontal distance between the abnormal node N23 and the normal nodes N21 and N22 may also be taken into consideration. This completes the layover correction step S32.

[0089] 10A and 10B are explanatory diagrams of the shadowing correction step S33. The larger the local incident angle, the more likely it is that a radar shadow has occurred. Therefore, the correction unit 33 determines that a radar shadow has occurred for node N2 included in an area where the local incident angle is equal to or greater than a third predetermined value (e.g., 90 degrees), and performs correction. Here, the third predetermined value is greater than the second predetermined value, which is the boundary value at which layover occurs. FIG. 10A is a map showing the local incident angle of each point in the form of contour lines. In FIG. 10A, areas where the local incident angle is equal to or greater than 90 degrees are shaded.

[0090] 10B is a graph showing how the elevation value of node N2 is corrected in the shadowing correction step S33. In FIG. 10B, node N2 at a point where the local incident angle is 90 degrees or greater is shown as "abnormal node N26." FIG. 10B includes an area where three consecutive abnormal nodes N26 occur.

[0091] Furthermore, a node N2 adjacent to the abnormal node N26 via the link L1 and having a local incident angle less than the third predetermined value (i.e., a node unlikely to be affected by shadowing) is referred to as a "normal node." In FIG. 10B, these nodes are shown as normal nodes N24 and N25. The normal node N24 is a node adjacent to the abnormal node N26 on one side. The normal node N25 is a node adjacent to the abnormal node N26 on the other side. In this embodiment, the node N2 whose local incident angle is equal to the third predetermined value is determined to be the abnormal node N26, but the node N2 may also be determined to be a normal node. That is, the correction unit 33 may determine a node N2 whose local incident angle exceeds the third predetermined value as an abnormal node, and a node N2 adjacent to an abnormal node and having a local incident angle equal to or less than the third predetermined value as a normal node.

[0092] Furthermore, the node N2 corresponding to the abnormal node N23, which is likely to have caused a layover, may be excluded from the normal nodes N24 and N25. That is, the normal nodes N24 and N25 are adjacent to the abnormal node N26 and correspond to a point where the local incident angle exceeds a second predetermined value and is less than a third predetermined value, which will be described later.

[0093] The correction unit 33 corrects the elevation value of the abnormal node N26 based on the elevation values ​​of the normal nodes N24 and N25. For example, the correction unit 33 changes the elevation value of the abnormal node N26 to the average value of the elevation value of the normal node N24 and the elevation value of the normal node N25. As in the elevation correction step S44 described above, the horizontal distance between the abnormal node N26 and the normal nodes N24 and N25 may also be taken into consideration when correcting the elevation value of the abnormal node N26. This completes the shadowing correction step S33.

[0094] In the layover correction step S32 and the shadowing correction step S33, the elevation value of the abnormal node is corrected based on the elevation value of the normal node adjacent to the abnormal node. The gradient of the road on which a vehicle travels usually does not change suddenly. Therefore, by correcting the elevation value of a node (abnormal node) for which an accurate elevation value is likely to be inaccurate due to a layover or radar shadow, based on the elevation value of a node (normal node) adjacent to the node and for which an accurate elevation value is likely to be obtained, a more accurate elevation value can be obtained.

[0095] By correcting the elevation value of node N2 through the elevated road correction step S31, layover correction step S32, and shadowing correction step S33, a node N3 including the corrected elevation value is generated. As a result, corrected data D4 (i.e., corrected road elevation data) including multiple nodes N3 and multiple links L1 is generated. Finally, the correction unit 33 stores the corrected data D4 in the second road elevation database 24. This completes the correction step S13.

[0096] Next, the road gradient data generation unit 34 performs a road gradient data generation step S14 (the "second generation step" of the present disclosure). First, the road gradient data generation unit 34 reads out the correction data D4 from the second road elevation database 24. Next, the road gradient data generation unit 34 calculates the gradient g2 based on the correction data D4.

[0097] The method for calculating gradient g2 is the same as the method for calculating gradient g1 in the gradient calculation step S41 described above. The difference is that gradient g1 is calculated based on node N2 before the elevation value is corrected, whereas gradient g2 is calculated based on node N3 after the elevation value is corrected. Specifically, the road gradient data generation unit 34 calculates gradient g2 between two nodes N3 located at the start and end points of link L1. The road gradient data generation unit 34 then generates link L2 by adding the calculated gradient g2 to the information on the corresponding link L1. As a result, road gradient data D5 including multiple nodes N3 and multiple links L2 is generated, as shown in FIG. 4(E). Finally, the road gradient data generation unit 34 stores the road gradient data D5 in the road gradient database 25. This completes the gradient data generation step S14.

[0098] As described above, the data generating device 30 generates road elevation data D3 based on elevation data D1 generated from SAR data and road map data D2, generates corrected data D4 by correcting the elevation value of the road elevation data D3, and generates road gradient data D5 by calculating gradient g2 based on the corrected data D4. The road gradient data D5 is used in the route search process of the route searching device 40, which will be described next.

[0099] <<About the route search process>> Link L1 in the road map data D2 includes information about the link cost LC as described above, and the information about the link cost LC is also inherited as is for link L2 in the road gradient data D5. Link cost LC is a value that represents the burden placed on a vehicle when traveling through link L1, and is calculated from information such as link travel time LT and link distance LD. Link travel time LT is, for example, the time required from entering the start point of link L1 to exiting the end point of link L1 and entering the start point of the next connecting link L1. Link distance LD is, for example, the distance between the start point and end point of link L1.

[0100] See Fig. 1. The route search device 40 searches for a route with the smallest travel cost (accumulated link costs LC) using a search algorithm such as the Dijkstra algorithm or the potential method. The Dijkstra algorithm is a search algorithm that, when constructing a tree of intermediate links from a starting link, and when a certain intermediate link branches off to another intermediate link, compares the magnitude of the route costs including the intermediate link after the branch (accumulated link costs LC from the starting link to the intermediate link after the branch), sorts the route costs in ascending order, and continues searching from the intermediate link with the smallest route cost.

[0101] The route search device 40 receives a search request from the electric vehicle EV1. The search request includes information about the starting point and the destination point. The route search device 40 searches for a route suitable for traveling by the electric vehicle EV1 based on the search request and the link cost LC included in the road gradient data D5.

[0102] Next, the route search device 40 predicts the power consumption of the electric vehicle EV1 at each point on the searched route based on the link cost LC and the gradient g2 included in the road gradient data D5. In particular, because the power consumption of the electric vehicle EV1 is significantly affected by the gradient of the road, incorporating the gradient g2 into the power consumption prediction makes it possible to more accurately predict the power consumption.

[0103] The route search device 40 then predicts the remaining battery charge of the electric vehicle EV1 at each point along the searched route by subtracting the predicted power consumption from the remaining battery charge of the electric vehicle EV1 at the departure point. This allows the route search device 40 to present a timing at which the electric vehicle EV1 should stop at a charging spot along the searched route. If necessary, the route search device 40 changes the searched route to a route that allows the electric vehicle EV1 to stop at a charging spot at an appropriate timing. This can alleviate the passenger's anxiety that the battery charge of the electric vehicle EV1 will run out on the way to the destination.

[0104] As described above, the data generating device 30 can provide the route searching device 40 with data (road gradient data) that can guide a more suitable route when searching for a route.

[0105] <<Variation>> Modifications of the embodiment will be described below. In the modifications, parts that are unchanged from the embodiment will be assigned the same reference numerals and descriptions thereof will be omitted.

[0106] <<Modification 1 of the elevation data generation unit>> In the above embodiment, the altitude data generator 31 generates the altitude data D1 based on SAR data. Since the SAR data is based on radio waves emitted from a satellite and reflected by the Earth's surface, it is affected by the weather and temperature on the Earth's surface. For example, thick clouds may reflect some of the radio waves. This can cause variations in the SAR data depending on the day the SAR data is acquired.

[0107] Therefore, the elevation data generation unit 31 may generate multiple pieces of elevation data D1 using SAR data observed at different dates and times as the original data, and use a statistical value (e.g., median, average, or mode) of the multiple pieces of elevation data D1 as the true value of the elevation data D1. For example, the elevation data generation unit 31 generates first elevation data based on first SAR data observed at a first date and time, and generates second elevation data based on second SAR data observed at a second date and time. Then, the elevation data generation unit 31 generates the elevation data D1 as the average value of the first elevation data and the second elevation data. With this configuration, it is possible to average out the effects of weather and the like, and to obtain more accurate elevation data D1.

[0108] <<Modification 2 of the Elevation Data Generation Unit>> In the above embodiment, the elevation data generation unit 31 uses DEM data to convert the absolute value of the phase after the offset into an elevation value (z) on the ground surface (phase-height conversion step S25). However, since the data generation device 30 is ultimately required to calculate the road gradient g2, it is not essential that the elevation value of the elevation data D1 represent an elevation value on the ground surface (i.e., how many meters above sea level) as long as the relative elevation value between points is accurate. For this reason, the phase-height conversion step S25 may be omitted, and the absolute value of the phase after the offset may be used directly as the elevation value of the elevation data D1. In this case, the number of steps in the elevation data generation unit 31 can be reduced, and DEM data becomes unnecessary.

[0109] For example, suppose the absolute value of the phase after offset at point A is 200 m, and the actual location is 100 m above sea level. Also, suppose the absolute value of the phase after offset at point B is 250 m, and the actual location is 150 m above sea level. In this case, if the phase-to-height conversion step S25 is omitted, the elevation values ​​of points A and B will be 200 m and 250 m, respectively, which are deviated by 100 m from the actual elevation values. However, because the difference in elevation values ​​between points A and B is 50 m in both cases, the gradient g2 will be the same value whether the phase-to-height conversion step S25 is executed or omitted.

[0110] <<Modification of the Road Elevation Data Generator>> In the above embodiment, the road elevation data generation unit 32 generates road elevation data D3 based on the elevation data D1 and the road map data D2. The elevation values ​​of the elevation data D1 may contain noise due to weather or other factors, as described above. For example, there may be a case where the elevation value suddenly increases by approximately 2 meters at a specific point. Such a point is unlikely to exist on a road where vehicles are normally allowed to travel, and this is likely to be noise. Therefore, the elevation values ​​of the road elevation data D3 may be corrected using a moving average to smooth the elevation values ​​at each point and remove noise.

[0111] For example, the average value (z31+z32+z33) / 3 of the elevation value z31 of a given node N2 and the elevation values ​​z32 and z33 of two nodes N2 adjacent to the given node N2 on one side and the other side via link L1 is set as the corrected elevation value of the given node N2. This makes it possible to remove the influence of noise and obtain more accurate road elevation data D3.

[0112] <<Modification 1 of the correction unit>> In the above embodiment, after extracting the link L1 whose gradient g1 is equal to or greater than the first predetermined value in the extraction step S42, the correction unit 33 determines whether the road corresponding to the link L1 is an elevated road by the operator's visual inspection or based on the road type of the link L1. However, while determining by the operator's visual inspection can provide a more reliable determination, it may increase the operator's workload. Also, there are areas where the link L1 is not assigned a road type indicating whether it is an elevated road or not. Therefore, the correction unit 33 according to the modified example determines whether the road corresponding to the link L1 is an elevated road based on an analysis of the reflection intensity image.

[0113] Here, a reflection intensity image is an image in SAR data that shows the strength of waves reflected from the Earth's surface. The strength of the reflected waves is expressed by pixel values. For example, in a reflection intensity image, areas with low reflection are displayed in black, and areas with high reflection are displayed in white. The strength of the reflected waves depends on the topography and the condition of the Earth's surface. For example, on smooth surfaces such as water or elevated roads, the radio waves emitted from the satellite are mostly reflected specularly, so almost no radio waves return to the satellite, and the strength of the reflected waves tends to be weak (black in the image). On the other hand, on rough surfaces such as forests, the radio waves emitted from the satellite are scattered, so some of the radio waves return to the satellite, and the strength of the reflected waves tends to be strong (white in the image).

[0114] Therefore, after extracting a link L1 whose gradient g1 is equal to or greater than a first predetermined value in the extraction step S42, the correction unit 33 determines whether or not the link L1 is an elevated road based on the reflection intensity at a point corresponding to the link L1 in the reflection intensity image. Note that the correction unit 33 may also include other points located within a predetermined distance range (for example, within a 100-meter square range) from the point corresponding to the link L1 as the basis for the determination, or may make the determination based only on the reflection intensity at those other points. If the correction unit 33 determines that the link L1 is an elevated road, it proceeds to the elevation correction step S44. If the correction unit 33 determines that the link L1 is not an elevated road, it ends the elevated road correction step S31.

[0115] In this way, by determining whether the road corresponding to link L1 is an elevated road based on the reflection intensity image, the operator confirmation step S43 can be omitted, and the altitude value can be corrected more quickly and easily.

[0116] <<Modification 2 of the correction unit>> In the above embodiment, the correction unit 33 corrects the elevation value for the node N2, which is likely to be an elevated road but for which the elevation value for the elevated road has not been acquired, in the elevated road correction step S31. Here, since the SAR data is data based on reflection from the ground surface, it is not possible to acquire the elevation value of a tunnel road. For this reason, even if the road type of the link L1 is a tunnel road, it is necessary to correct the elevation value.

[0117] The correction unit 33 of this modification further performs a tunnel road correction step S34 in addition to the elevated road correction step S31. The tunnel road correction step S34 is a step executed in the correction step S13. The tunnel road correction step S34 may be executed before or after the elevated road correction step S31.

[0118] 11 is a flowchart showing a subroutine of the tunnel road correction step S34. First, the correction unit 33 reads the road elevation data D3 from the first road elevation database 23 and determines whether the road type of the link L1 is a tunnel road (tunnel road determination step S51). If the road type of the link L1 is a tunnel road (YES route in S51 in FIG. 11), the process proceeds to the elevation correction step S52. If the road type of the link L1 is other than a tunnel road (NO route in S51 in FIG. 11), the tunnel road correction step S34 is terminated.

[0119] Figure 12 is a graph illustrating the elevation correction step S52. The vertical axis of Figure 12 represents elevation value, and the horizontal axis represents horizontal distance. The left side of the horizontal axis is referred to as the first side, and the right side is referred to as the second side. In Figure 12, link L1 whose road type is a tunnel road is shown as a "tunnel link L31," and link L1 whose road type is other than a tunnel road is shown as a "non-tunnel link L32."

[0120] Node N2, which connects to tunnel link L31 on both the first and second sides, is referred to as "abnormal node N29." Abnormal node N29 is a node N2 set within the tunnel road because both the incoming link L1 and the outgoing link L1 are tunnel links L31. For this reason, the elevation value of abnormal node N29 is not the actual elevation value of the tunnel road, but the elevation value of the surface of the mountain through which the tunnel road passes, and so a correction of the elevation value is required.

[0121] Here, a node N2 having one of the first and second sides connected to the tunnel link L31 and the other of the first and second sides connected to the non-tunnel link L32 is referred to as a "normal node." In FIG. 12, normal node N27 is a node N2 having a second side connected to the tunnel link L31 and a first side connected to the non-tunnel link L32, and normal node N28 is a node N2 having a first side connected to the tunnel link L31 and a second side connected to the non-tunnel link L32. In this way, node N2 located at the boundary between tunnel link L31 and non-tunnel link L32 is considered to be a node N2 set at the entrance or exit of a tunnel road (or in the vicinity of these). Since it is considered that correct elevation values ​​are obtained from SAR data up until just before entering the tunnel road, the elevation value of abnormal node N29 is corrected based on the elevation values ​​of normal nodes N27 and N28.

[0122] Specifically, the correction unit 33 changes the elevation value of the abnormal node N29 to the average value of the normal nodes N27 and N28. As in the above-mentioned elevation correction step S44, when correcting the elevation value of the abnormal node N29, the horizontal distance between the abnormal node N29 and the normal nodes N27 and N28 may also be taken into consideration. By correcting the elevation value of the abnormal node N29 in this way, an accurate elevation value can be obtained. This completes the elevation correction step S52.

[0123] Depending on the resolution of the SAR data, there is a risk that the correct elevation value cannot be obtained immediately before entering a tunnel road. For this reason, the node N2 that satisfies the above conditions for becoming the normal nodes N27 and N28 may be set as the abnormal node N29 that requires correction, and the node N2 that is connected to the non-tunnel link L32 on both the first and second sides and is adjacent to the abnormal node N29 via the non-tunnel link L32 may be set as the "normal node." In this case, the nodes N27a and N28a shown in Figure 12 are normal nodes.

[0124] "others" The road gradient data generator 34 according to the above embodiment calculates the gradient g2 between two nodes N3 located at the start and end points of a single link L1. However, the gradient g2 may be the gradient between any two nodes N3, and the number of links L1 between the nodes N3 may be two or more. For example, the gradient g2 may be calculated as the gradient between the node N3 located most upstream of a plurality of serially connected links L1 (a group of links) (i.e., the start point of a group of serially connected links) and the node N3 located most downstream of the plurality of links L1 (i.e., the end point of a group of serially connected links).

[0125] In the road gradient data D5 according to the above embodiment, the gradient g2 is stored as information about the link L2. However, the method for storing the gradient g2 in the road gradient data D5 is not limited. For example, the gradient g2 may be assigned as coordinate information (x, y, z, g2) of the node N3. Alternatively, the gradient g2 may be stored in the road gradient data D5 as independent information without being assigned to either the link L2 or the node N3.

[0126] The gradient information in the above embodiment is the gradient g2 between two nodes N3. However, the gradient information may be any information that represents the inclination of the road, and may be a numerical value other than the gradient, or may be information that represents the degree of the gradient. For example, the gradient may be divided into predetermined numerical ranges, and information such as "large gradient," "small gradient," or "no gradient" may be used as the gradient information. When calculating the remaining battery charge of the electric vehicle EV1, the required calculation accuracy can be ensured if an approximate gradient degree is obtained. When a specific gradient value is not required, it is preferable to use the gradient information as information that represents the degree of the gradient in order to reduce the amount of data.

[0127] The elevation data D1 according to the above embodiment is generated based on SAR data. However, the elevation data D1 may be generated based on data other than SAR data. For example, the elevation data D1 may be generated based on data acquired by using an aircraft other than a satellite, such as an unmanned aircraft, to emit radio waves to the Earth's surface and receive the waves reflected from the Earth's surface.

[0128] 《Addendum》 It should be noted that at least some of the above-described embodiments and various modified examples may be combined with each other in any desired manner. Furthermore, the embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, and it is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0129] 100 Route Search System 10. Communications Department 20 databases 21 Elevation Database 22 Road Map Database 23 First Road Elevation Database 24 Second Road Elevation Database 25 Road gradient database 30 Data generation device 31 Elevation data generation unit 32 Road elevation data generation unit 33 Correction unit 34 Road gradient data generation unit 40 Route search device 50 satellite databases 61 Display section 62 Input section D1 elevation data D2 Road map data D3 Road elevation data D4 Corrected data (corrected road elevation data) D5 Road gradient data NT1 communication network EV1 Electric Vehicle N1 node N2 (altitude value assigned) node N3 (Elevation value corrected) node N11, N12 normal nodes (first normal nodes) N13 Abnormal node (first abnormal node) N21, N22 normal nodes (second normal nodes) N23 Abnormal node (second abnormal node) N24, N25 normal nodes (third normal node) N26 Abnormal node (third abnormal node) N27, N28 normal nodes (fourth normal node) N27a, N28a nodes (fourth normal node) N29 Abnormal Node (4th abnormal node) L1 Link L2 (graded) link L31 Tunnel Link L32 non-tunnel link LC Link Cost LT Link Travel Time LD link distance g1 Gradient (between node N2) g2 Gradient (between node N3) AR1, AR2, AR3 area

Claims

1. a first generating unit that generates road elevation data in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes, based on elevation data including elevation values ​​of the plurality of points and road map data in which roads are represented by a plurality of nodes and a plurality of links; a second generating unit that calculates gradient information between any two of the plurality of nodes based on the road elevation data, and generates road gradient data by assigning the gradient information to the plurality of nodes or the plurality of links; a third generation unit that generates the altitude data based on SAR data obtained by a synthetic aperture radar; a correction unit that corrects the altitude value assigned to the node, the correction unit corrects the elevation value of the abnormal node, which is the node that satisfies the correction condition, based on the elevation value of a normal node, which is the node that is adjacent to the abnormal node and does not satisfy the correction condition; The second generation unit generates the road gradient data based on the corrected road elevation data.

2. the third generation unit generates a plurality of pieces of elevation data based on the SAR data observed at different dates and times, and defines a statistical value of the generated plurality of pieces of elevation data as a true value of the elevation data. The data generating device according to claim 1 .

3. the abnormal node includes a first abnormal node connected to a target link, the link corresponding to an elevated road and the link having a gradient between a start point and an end point that exceeds a first predetermined value; the normal nodes include a first normal node connected to an asymmetric link other than the target link and adjacent to the first abnormal node; the correction unit corrects the elevation value of the first abnormal node based on the elevation value of the first normal node. The data generating device according to claim 1 .

4. the correction unit determines whether the link corresponds to an elevated road based on the reflection intensity of at least one of a point corresponding to the link and a point located within a predetermined distance range from the point in a reflection intensity image that represents the strength of a reflected wave returning from the Earth's surface to a satellite; The data generating device according to claim 1 .

5. the anomaly nodes include a second anomaly node corresponding to a point in the SAR data where the local incidence angle is less than a second predetermined value; the normal nodes include a second normal node adjacent to the second abnormal node and corresponding to a point where the local incident angle exceeds the second predetermined value; the correction unit corrects the elevation value of the second abnormal node based on the elevation value of the second normal node. The data generating device according to any one of claims 1 to 4.

6. the anomaly nodes include a third anomaly node corresponding to a point in the SAR data where a local angle of incidence exceeds a third predetermined value; the normal nodes include a third normal node adjacent to the third abnormal node and corresponding to a point where the local incident angle is less than the third predetermined value; the correction unit corrects the elevation value of the third abnormal node based on the elevation value of the third normal node. The data generating device according to any one of claims 1 to 4.

7. The abnormal node is a second anomaly node corresponding to a point where a local angle of incidence in the SAR data is less than a second predetermined value at which a layover occurs in the SAR data; a third anomaly node corresponding to a point where the local angle of incidence exceeds a third predetermined value at which a radar shadow appears in the SAR data; the third predetermined value is greater than the second predetermined value, The normal node is a second normal node adjacent to the second abnormal node, the second normal node corresponding to a point where the local incident angle exceeds the second predetermined value and is less than a third predetermined value; a third normal node adjacent to the third abnormal node and corresponding to a point where the local incident angle exceeds the second predetermined value and is less than a third predetermined value; The correction unit correcting the elevation value of the second abnormal node based on the elevation value of the second normal node; correcting the elevation value of the third abnormal node based on the elevation value of the third normal node; The data generating device according to any one of claims 1 to 4.

8. the abnormal nodes include a fourth abnormal node connected to a tunnel link, which is the link corresponding to a tunnel road; the normal nodes include a fourth normal node connected to a non-tunnel link that is the link adjacent to the fourth abnormal node and corresponds to a road other than the tunnel road; the correction unit corrects the elevation value of the fourth abnormal node based on the elevation value of the fourth normal node. The data generating device according to any one of claims 1 to 7.

9. A computer-implemented data generation method, comprising: a first generation step of generating road elevation data in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes, based on elevation data including elevation values ​​of the plurality of points and road map data in which roads are represented by a plurality of nodes and a plurality of links; a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data by assigning the gradient information to the plurality of nodes or the plurality of links; a third generation step of generating the altitude data based on SAR data obtained by a synthetic aperture radar; a correction step of correcting the altitude value assigned to the node, the correction step includes a process of correcting an elevation value of an abnormal node, which is the node that satisfies a correction condition, based on an elevation value of a normal node, which is the node that is adjacent to the abnormal node and does not satisfy the correction condition; The data generating method, wherein the second generating step includes generating the road gradient data based on the corrected road elevation data.

10. A computer program for operating a computer as a data generating device, a first generation step of generating road elevation data in which the elevation values ​​of the points corresponding to the plurality of nodes are assigned to the nodes, based on elevation data including elevation values ​​of the plurality of points and road map data in which roads are represented by a plurality of nodes and a plurality of links; a second generation step of calculating gradient information between any two of the plurality of nodes based on the road elevation data, and generating road gradient data by assigning the gradient information to the plurality of nodes or the plurality of links; a third generation step of generating the altitude data based on SAR data obtained by a synthetic aperture radar; a correction step of correcting the altitude value assigned to the node, the correction step includes correcting an elevation value of an abnormal node, which is the node that satisfies a correction condition, based on an elevation value of a normal node, which is the node that is adjacent to the abnormal node and does not satisfy the correction condition; The second generating step includes generating the road gradient data based on the corrected road elevation data.

Images