JP2023106516A - Feature point generation method - Google Patents

Abstract

To provide a feature point generation method, a feature point generation device and a feature point generation program with which it is possible to generate the feature points of a travel path, and a recording medium in which the feature point generation program is recorded.SOLUTION: A server generates a feature point that constitutes a candidate for a travel path, on the basis of the travel trajectory of a moving body. In this case, the server acquires the travel trajectory from an onboard device via a network. The server extracts a linear component from the travel trajectory, finds the intersection of an extension of the extracted linear component and extensions of other linear components adjacent to the linear component in the travel trajectory, and generates this intersection as a feature point when the travel trajectory passes through within a prescribed range from this intersection.SELECTED DRAWING: Figure 3

Description

The present invention relates to a feature point generation method.

When a vehicle travels on a new road that does not exist on a map represented by existing map data, such as a newly opened road, the map data is updated by estimating the shape of the new road from the vehicle's position information and driving direction. A technique for doing so is known (Patent Document 1).

When the above technology is used on roads with many intersections and turns, such as roads in parking lots, the generation of feature points for identifying intersections, turns, etc., in addition to the shape, is useful for navigation and autonomous driving. essential for use. However, in the technique of Patent Document 1, as an example, there is a problem that new feature points are not generated only by estimating the shape of the new traveling road.

JP 2014-235510 A

An object of the present invention is to address such problems. That is, it is an object of the present invention to provide, for example, a feature point generation method, a feature point generation device, a feature point generation program, and a recording medium recording the feature point generation program, which can generate feature points of a road. and

A feature point generation method according to claim 1, which has been made to solve the above-described problems, is a feature point generation method for generating feature points that are candidates for nodes of a travel path based on a travel trajectory of a moving object, an acquisition step of acquiring the running locus, an extracting step of extracting a straight line component from the running locus, an extension line of the extracted straight line component, and an extension line of another straight line component adjacent to the straight line component in the running locus and a first generation step of generating intersections of and as the feature points.

Further, the feature point generation device according to claim 5 is a feature point generation device that generates feature points that are candidates for nodes of a travel path based on the travel trajectory of a moving body, and includes an acquisition unit that acquires the travel trajectory. and an extraction unit that extracts a straight line component from the running locus, an extension line of the extracted straight line component, and an extension line of another straight line component adjacent to the straight line component in the running locus, the intersection point being the feature point and a first generator that generates as .

Further, the feature point generation program according to claim 6 is a feature point generation program that causes a computer to generate feature points that are candidate nodes of a travel path based on a travel locus of a moving body, wherein the computer includes: an acquisition unit that acquires a running locus, an extracting unit that extracts a straight line component from the running locus, an extension line of the extracted straight line component, and an extension line of another straight line component adjacent to the straight line component in the running locus , as the feature point.

A recording medium according to claim 7 is characterized by recording the feature point generating program according to claim 7.

1 shows a system configuration using a feature point generation device and a link information generation device according to an embodiment of the present invention; 2 is a flow chart of operations related to feature point generation of the feature point generation device shown in FIG. 1; FIG. 4 is an explanatory diagram showing a specific example of feature point generation; FIG. 4 is an explanatory diagram showing a specific example of feature point generation; FIG. 4 is an explanatory diagram showing a specific example of feature point generation; FIG. 11 is an explanatory diagram showing a modified example of feature point generation; FIG. 4 is an explanatory diagram showing a specific example of feature point integration; 2 is a flow chart of the operation of the feature point generation device shown in FIG. 1; (A) shows the result of integration when there are few running trajectories, and (B) shows the result of merging when there are many running trajectories. FIG. 4 is an explanatory diagram showing a specific example of link information generation; FIG. 4 is an explanatory diagram showing a specific example of link information generation; FIG. 2 is a flowchart of operations of the link information generating device shown in FIG. 1; FIG. FIG. 4 is an explanatory diagram showing a specific example of link information generation; FIG. 4 is an explanatory diagram showing a specific example of link information generation;

A feature point generation method according to an embodiment of the present invention is a feature point generation method for generating feature points that are candidate nodes of a travel path based on a travel trajectory of a moving body, and is a method for acquiring the travel trajectory. an extraction step of extracting a straight line component from the running locus; an extension line of the extracted straight line component and an extension line of another straight line component adjacent to the straight line component in the running locus; and a first generating step of generating as points. As a result, it is possible to generate characteristic points of a travel path that are suitable for use in navigation and automatic driving.

Further, when the running locus does not exist within a predetermined range centered on the intersection, or when the intersection does not exist, a locus from a point on the running locus that deviates from the straight line component to the other straight line component. A second generation step of generating, as the characteristic points, an intersection point between a tangent line on the top and an extension line of the straight line component, and an intersection point between the tangent line and an extension line of the other straight line component. This makes it possible to generate feature points that are even more suitable for use in navigation and automatic driving.

Also, the tangent line may be a tangent line at the center of the trajectory.

Further, the method may include a third generation step of generating, as the feature point, a point where the change in the inclination angle of the travel locus is equal to or greater than a predetermined angle. This makes it possible to generate feature points that are even more suitable for use in navigation and automatic driving.

Further, a feature point generation device according to an embodiment of the present invention is a feature point generation device that generates feature points that are candidates for nodes of a travel path based on a travel trajectory of a moving object, and obtains the travel trajectory. an acquisition unit that extracts a straight line component from the running locus, an extension line of the extracted straight line component, and an extension line of another straight line component that is adjacent to the straight line component in the running locus. and a first generator for generating the feature points. Thereby, feature points of the new road can be generated.

Moreover, it is good also as a feature point generation program which makes a computer perform the feature point generation method mentioned above. Since the program is executed by a computer in this way, dedicated hardware or the like is not required, and it can be installed and functioned in a general-purpose information processing apparatus.

Also, the feature point generation program described above may be stored in a computer-readable recording medium. By doing so, the program can be distributed as a single unit in addition to being incorporated into the device, and version upgrades and the like can be easily performed.

A feature point generation device and a link information generation device of the present invention will be described with reference to FIGS. 1 to 14. FIG. A server 10 as a feature point generation device and a link information generation device according to an embodiment of the present invention includes a communication unit 11, a feature point generation unit 12, a link information generation unit 13, and a storage unit 14. there is

The communication unit 11 communicates with the vehicle-mounted device 20 to be described later via a network N such as the Internet. The communication unit 11 receives, for example, the position information (latitude and longitude) of the vehicle (moving body) acquired by the vehicle-mounted device 20, the traveling direction, and the traveling locus such as the inclination angle. Further, when the in-vehicle device 20 is a device that requires map data, such as a navigation device, the communication unit 11 downloads map data stored in the storage unit 14 to be described later in response to a request from the in-vehicle device 20. Send.

The feature point generation unit 12 generates feature points that are candidate nodes on the travel road based on the travel locus received from the communication unit 11 .

The link information generation unit 13 generates links connecting the feature points generated by the feature point generation unit 12 .

The storage unit 14 stores the feature points generated by the feature point generation unit 12 and the links generated by the link information generation unit 13 . In addition, the storage unit 14 accumulates and stores travel loci received from the plurality of vehicle-mounted devices 20 . The storage unit 14 also stores existing map data in which the vehicle travel route is indicated by nodes, links, and the like.

In the above description, the feature point generation unit 12 and the link information generation unit 13 function as an arithmetic device such as a CPU (Central Processing Unit) of the server 10 . Also, the network interface of the server 10 and the like function as the communication unit 11 . A storage medium such as a hard disk of the server 10 functions as the storage unit 14 . 2, 8, and 12 executed by the feature point generation unit 12 and the link information generation unit 13 are configured as computer programs executed by the CPU, the feature point generation program and the link information generation It can be a program. Further, the feature point generation unit 12 and the link information generation unit 13 execute the flowcharts shown in FIGS. 2, 8, and 12, so that the server 10 implements the feature point generation method and the link information generation method of the present invention. .

The in-vehicle device 20 is mounted on a vehicle, and collects and transmits travel loci to the server 10 . Further, when the vehicle-mounted device 20 is configured as a navigation device, the vehicle-mounted device 20 requests generated map data from the server 10 as necessary.

The in-vehicle device 20 includes a communication section 21 , a control section 22 and a vehicle information acquisition section 23 .

The communication unit 21 communicates with the server 10 via a network N such as the Internet. The communication unit 21 transmits the travel locus including the position information (latitude and longitude) of the vehicle acquired by the vehicle information acquisition unit 23, the traveling direction, the inclination angle, and the like. In this embodiment, the travel locus is composed of point cloud data representing positions (latitude and longitude) on the locus along which the vehicle has traveled, as shown in FIG. 3, for example. , angle of inclination, etc. are added.

Further, when the in-vehicle device 20 is a device such as a navigation device that requires map data, the communication unit 21 requests map data from the server 10 under the control of the control unit 22 or the like, and transmits the map data from the server 10 . receive map data.

The control unit 22 manages overall control of the vehicle-mounted device 20 . The control unit 22 causes the communication unit 21 to transmit the position information (latitude, longitude), traveling direction, inclination angle, etc. of the vehicle collected by the vehicle information acquisition unit 23 to the server 10 as a travel locus.

The vehicle information acquisition unit 23 is connected to various sensors and receivers such as a GPS (Global Positioning System) receiver, a gyro sensor, and a vehicle speed pulse. The vehicle information acquisition unit 23 collects information such as position information (latitude and longitude), direction of travel, and angle of inclination of both vehicles from the above-described sensors and the like.

Note that the in-vehicle device 20 is not limited to a car navigation system installed in an instrument panel or the like of a vehicle, and may be a portable device such as a smart phone. That is, the in-vehicle device 20 may be detachably installed in the vehicle.

In the above description, the controller 22 functions as an arithmetic device such as a CPU of the vehicle-mounted device 20 . In addition, a communication circuit, an antenna, and the like of the vehicle-mounted device 20 function as the communication unit 21 . In addition, the vehicle information acquisition unit 23 functions as an interface circuit with various sensors.

Next, operations of the server 10 and the vehicle-mounted device 20 configured as described above will be described. The in-vehicle device 20 transmits the travel locus to the server 10 at predetermined timings such as every predetermined time period or every time the vehicle travels a predetermined distance. The server 10 functions as an acquisition unit, receives (acquires) the travel locus from the in-vehicle device 20 , and sequentially stores it in the storage unit 14 .

The above-described server 10 (feature point generation unit 12) executes feature point generation processing shown in FIG. The server 10 reads out one of the plurality of running trajectories stored in the storage unit 14 for which the feature point generating process has not been executed (step S1). Subsequently, the server 10 uses a well-known map matching technique to determine whether or not the read travel locus is a travel locus that deviates from the travel route included in the map data (step S2).

When the server 10 determines that the travel locus is the travel route included in the map data (N in step S2), the process ends.

On the other hand, when the server 10 determines that the traveled locus deviates from the traveled route included in the map data (Y in step S2), the server 10 advances to steps S3 to S13 to select candidate nodes of the traveled route from the traveled locus. generate feature points.

First, the server 10 works as an extraction unit and extracts a straight line component from the travel locus using a well-known straight line extraction technique (step S3). Next, as shown in FIG. 3A, the server 10 connects an extension line LPn of the extracted linear component LSn and an extension line LPn+1 of another linear component LSn+1 adjacent to the linear component LSn on the travel locus. Find the intersection (step S4). If the intersection can be found in step S4 (Y in step S5), the server 10 sets the intersection found in step S4 as a feature point candidate Pcn as shown in FIG. 3B (step S6).

Next, the server 10 determines whether or not the travel locus exists within a predetermined range A1 centering on the feature point candidate Pcn (step S7). In this embodiment, the predetermined range A1 is an area within a circle centered on the feature point candidate Pcn, but is not limited to this. The predetermined range A1 may have any shape as long as it is centered on the feature point candidate Pcn.

As shown in FIG. 3(B), if the running locus exists within a predetermined range A1 centering on the feature point candidate Pcn (Y in step S7), the server 10, as shown in FIG. 3(C), After functioning as a first generator and determining the feature point candidate Pcn as the feature point Pfn (step S8), the process proceeds to step S13.

On the other hand, as shown in FIG. 4A, for example, when the radius of curvature of the running locus is large, the server 10 determines that the running locus does not exist within the predetermined range A1 centered on the feature point candidate Pcn (in step S5 N), go to step S9. Also, the server 10 proceeds to step S9 even when the intersection of the extension lines LPn and LPn+1 cannot be made (N in step S5), such as the trajectory of crank movement shown in FIG.

In step S9, as shown in FIGS. 4(B) and 5, the server 10 generates a tangential line at the center of the trajectory from the point Pn deviating from the straight line component LSn to the point Pn+1 on the other straight line component LSn+1 in the running trajectory. Find Ltn. Then, the server 10 sets the intersection point between the obtained tangent line Ltn and the extension line LPn of the straight line component LSn as a feature point candidate Pcn1, and sets the intersection point between the obtained tangent line Ltn and the extension line LPn+1 of the straight line component LSn+1 as a feature point candidate Pcn2. (step S10).

After that, the server 10 determines whether or not the travel locus exists within a predetermined range A1 centering on the feature point candidates Pcn1 and Pcn2 (step S11), as in step S7. When the running locus exists within the predetermined range A1 centered on the feature point candidates Pcn1 and Pcn2 (Y in step S11), the server 10 functions as a second generation unit and uses the feature point candidates Pcn1 and Pcn2 as feature points Pfn. After confirming, the process proceeds to step S13. On the other hand, if the running locus does not exist within the predetermined range A1 centered on the feature point candidates Pcn1 and Pcn2 (N in step S11), the server 10 proceeds to step S13 without determining the feature point Pfn.

After that, the server 10 determines whether or not the process of determining feature points in steps S4 to S12 has been executed for all adjacent straight line components in the running locus (step S13). If feature points have not been determined for all of the adjacent linear components (N in step S13), the process returns to step S4 to execute processing for determining feature points for the next adjacent linear components LSn+1 and LSn+2 on the running locus. .

According to the feature point generation process described above, the server 10 extracts a straight line component from the running locus, and extracts an extension line LPn of the extracted straight line component LSn and an extension of another straight line component LSn+1 adjacent to the straight line component LSn in the running locus. A feature point candidate Pcn that is an intersection of the line LPn+1 and the line LPn+1 is generated as a feature point Pfn. As a result, it is possible to generate feature points at positions that are candidates for nodes such as corners and intersections on the road, and it is possible to generate feature points of the road that are suitable for use in navigation and automatic driving.

Further, according to the feature point generation process described above, when the running locus does not exist within the predetermined range A1 around the feature point candidate Pcn, which is the intersection of the extension lines LPn and LPn+1, or when the extension lines LPn and LPn+1 does not exist, feature point candidates Pcn1 and Pcn2, which are the intersections of the tangent line Ltn and the extension lines LPn and LPn+1, are generated as feature points Pfn.

As a result, as shown in FIG. 4, when the radius of curvature of the traveled path is large, the feature point Pfn can be generated at a position near the bend angle of the traveled path and close to the traveled path. As a result, it is possible to perform map matching with high accuracy, and it is possible to generate characteristic points of the road that are even more suitable for use in navigation and automatic driving.

In the above embodiment, if the running locus does not exist within the predetermined range A1 centering on the intersection of the extension lines LPn and LPn+1, the intersection is not generated as the characteristic point Pfn, but the invention is not limited to this. If there is an intersection of the extension lines LPn and LPn+1, the intersection may be set as the characteristic point Pfn without determining whether or not the travel locus exists within the predetermined range A1.

Further, when the running locus does not exist within the predetermined range A1 centering on the intersection of the extension lines LPn and LPn+1, the feature point Pfn is determined based on the extension lines LPn and LPn+1 without obtaining new feature point candidates Pcn1 and Pcn2. It doesn't have to be generated.

Further, in the above-described embodiment, when the running locus does not exist within the predetermined range A1 centering on the intersection points (feature point candidates Pcn1 and Pcn2) between the tangent line Ltn and the extension lines LPn and LPn+1, the feature point Pfn is not generated. It is not limited to this. The intersection of the tangent line Ltn and the extension lines LPn and LPn+1 may be generated as the feature point Pfn without determining whether the running locus exists within the predetermined range A1 centered on the feature point candidates Pcn1 and Pcn2.

Further, when the running locus does not exist within the predetermined range A1 centered on the feature point candidates Pcn1 and Pcn2, tangent lines are drawn at positions slightly shifted from the center toward the points Pn and Pn+1, respectively, and the tangent line and the extension line LPn, The intersection with LPn+1 may be used as a new feature point candidate. Then, it is conceivable to shift the tangent line to the point Pn, Pn+1 side until the running locus exists within a predetermined range A1 centering on the new feature point candidate.

In addition, in the above-described embodiment, the feature point Pfn is generated from the position (latitude and longitude) of the running locus, but the present invention is not limited to this. For example, as shown in FIG. 6, the server 10 may set a feature point Pfn as a position on the travel locus where the change in inclination angle is greater than or equal to a predetermined angle. In this embodiment, the feature points Pfn and Pfn+1 are the positions on the travel locus where the change in the inclination angle is 1 degree or more. As a result, for example, feature points Pfn and Pfn+1 can be generated at the boundary between each floor of a self-driving multi-storey parking lot and the slope connecting the floors. Feature points may be generated using altitude data from GPS, but there is a high possibility that errors in altitude data from GPS are large. can get. Therefore, it is possible to generate feature points Pfn and Pfn+1 that are even more suitable for use in navigation and automatic driving.

Also, a camera is installed in the vehicle, and an image of the camera is transmitted from the vehicle to the server 10. - 特許庁If the server 10 can detect a landmark such as a toll gate by analyzing the camera image, the point may be generated as a feature point. Furthermore, a vehicle may be equipped with a narrowband communication device (for example, DSRC) that communicates with a tollgate or the like, and the location information transmitted to the server 10 may be added to indicate that communication has been performed. The server 10 generates the position at which communication is performed as a feature point.

By executing the feature point generation process described above, feature points can be generated for each of the plurality of travel trajectories. Therefore, as shown in FIG. 7, for example, a plurality of feature points Pf are generated for intersections on the same road.

Therefore, each time a feature point Pfn is determined by the feature point generation process (that is, each time steps S9 and S12 are executed), the server 10 functions as an integration unit and integrates a plurality of feature points Pfn. 8 is executed.

In the feature point integration process, the server 10 determines whether or not an existing feature point Pfn determined prior to the determined feature point Pfn exists within a predetermined range A2 centered on the determined feature point Pfn ( step S21). The predetermined range A2 may be the same as or different from the predetermined range A1 used in the feature point generation process. If the existing feature point Pfn exists within the predetermined range A2 (Y in step S21), the server 10 obtains the average position of the two feature points Pfn present within the predetermined range A2 (step S22). Then, the server 10 integrates the feature points into the average position (step S23).

After that, the server 10 deletes the two feature points Pfn before integration, saves the integrated feature point Pfn as an existing feature point Pfn in the storage unit 14 (step S24), and ends the process. On the other hand, if the existing feature point Pfn does not exist within the predetermined range A2 (N in step S21), the server 10 saves the feature point Pfn determined this time as an existing feature point in the storage unit 14 (step S24), the process ends.

According to the feature point integration process described above, when a plurality of generated feature points Pfn exist within the predetermined range A2, the plurality of feature points Pfn can be integrated. As a result, it is possible to generate feature points suitable for use in navigation and automatic driving without adding a plurality of feature points Pfn to one turn or intersection of the travel road.

Further, according to the feature point integration process described above, the positions (latitude and longitude) of the plurality of feature points Pfn are integrated into a feature point that is the average. This makes it possible to generate feature points Pfn that are even more suitable for use in navigation and automatic driving.

In the feature point integration processing described above, the integration result differs depending on the number of running loci. FIG. 9A shows the feature point integration result when there are few running trajectories. FIG. 9B shows the feature point integration result when there are many running trajectories. When there are many running trajectories, the position of feature point 1 moves a little lower than when there are few. In general, it is considered that the feature points are closer to the average of the roads on which the actual vehicle travels as the number of travel trajectories increases. In this way, by generating feature points according to the actual travel route and using the map data, it is possible to improve the accuracy of map matching.

Note that the above-described feature point integration processing integrates feature points each time one feature point is generated by the feature point generation processing, but the present invention is not limited to this. The feature point integration process may be performed after performing the feature point generation process for a plurality of running trajectories (that is, after performing the feature point generation process a plurality of times). In this case, similarly, the latitudes and longitudes of a plurality of feature points existing within the predetermined range A2 are averaged and integrated into the averaged point. Further, if a plurality of feature points existing within the predetermined range A2 have the same latitude and longitude, the feature points may be integrated into the latitude and longitude with the largest number. Furthermore, the feature points may be integrated at the central position of a graphic formed by connecting a plurality of feature points existing within the predetermined range A2.

In addition, the feature point integration process described above is a process of integrating the feature points generated from the extension lines LPn and LPn+1 of the running locus, but it is conceivable to similarly integrate the feature points generated from the inclination angle. For the feature point generated from the tilt angle, the tilt angle is added to the feature point. If there are a plurality of feature points having the same inclination angle within the predetermined range A2, the feature points may be integrated by averaging the latitude and longitude.

Next, the server 10 generates a link connecting the feature points integrated by the feature point integration process described above. Basically, the server 10 (the link information generation unit 13 thereof) generates characteristic points according to the passage order of one running locus R1 passing through a predetermined range A3 from each of the plurality of characteristic points Pf7 and Pf18 as shown in FIG. The points Pf7 and Pf18 are connected. The predetermined range A3 may be the same as or different from the predetermined ranges A1 and A2 used in the feature point generation process and the feature point integration process. At this time, the connection direction of the feature points (passing direction of the vehicle) may also be added to the link data.

However, in this link generation method, as shown in FIG. 11(B), when a link connecting the feature points Pf3 and Pf2 is generated along the running locus R2, a link is generated based on the running locus R1. Since the running locus R2 passes through the predetermined range A3 centered on the feature point Pf8, as shown in FIG. put away.

Therefore, in this embodiment, the link generation process shown in FIG. 12 is executed. First, the server 10 determines whether the same running locus passes within a predetermined range A3 from each feature point among a plurality of feature points generated in the feature point generation process and integrated in the feature point integration process, and Three consecutively ordered feature points are extracted (step S30). For example, in the example shown in FIG. 11, the feature points Pf2, Pf8, and Pf3 are three features in which the running locus R2 (one of a plurality of running loci) passes within the predetermined range A3 and the passage order is continuous. Extracted as points.

Next, the server 10 creates a line L1 connecting the extracted continuous three feature points Pf2, Pf8, and Pf3, and a running locus R2 passing through a predetermined range A3 from each of the three feature points Pf2, Pf8, and Pf3. , is calculated (step S31). In this embodiment, as shown in FIG. 11A, the server 10 first evaluates the maximum value of the distances until the normal line (indicated by the arrow in the drawing) on the running locus R2 intersects the line L1. result. In the example shown in FIG. 11A, the maximum value is the distance until the normal crosses the feature point Pf8.

Next, the server 10 calculates a second evaluation result by evaluating the distance between the line L2 connecting the first and last characteristic points Pf2, Pf3 of the three characteristic points Pf2, Pf8, Pf3 and the traveling locus R2. (step S32). In this embodiment, as shown in FIG. 11B, the server 10 takes the maximum value of the distances until the normal line on the running locus R2 intersects the line L2 as the second evaluation result. In the example shown in FIG. 11B, the maximum value is the distance until the normal crosses the feature point Pf2.

Next, the server 10 compares the first evaluation result and the second evaluation result, and if the first evaluation result is larger than the second evaluation result by a certain value or more (Y in step S33), A link connecting the front and rear feature points Pf2 and Pf3 is generated without connecting Pf8, and the process ends (step S34). On the other hand, if the first evaluation result is not larger than the second evaluation result by a certain value or more (N in step S33), the server 10 connects three consecutive feature points Pf2, Pf8, and Pf3, and ends the process. (Step S35).

Especially on the road in a parking lot, the direction of travel is often determined in detail. In the example shown in FIG. It was an inappropriate link. According to the embodiment described above, in the example shown in FIG. 11, a link connecting feature points Pf2 and Pf3 is generated. Therefore, the accuracy of the link is improved, and link information suitable for use in navigation and automatic driving can be generated.

In addition, the first evaluation result and the second evaluation result are not limited to the examples described above. As the first evaluation result, for example, as shown in FIG. It may be an area surrounded by a normal line L4 on the running locus R2 passing through (indicated by diagonal lines in the figure). For example, as shown in FIG. 13B, the second evaluation result may be an area (indicated by hatched lines in the drawing) surrounded by the running locus R2, the line L2, and the normal lines L3 and L4.

As the first evaluation result, as shown in FIG. 14A, the maximum distance between the normal line on the running locus R2 and each of the three feature points Pf2, Pf8, and Pf3 may be used. As the second evaluation result, as shown in FIG. 14B, the maximum value of the distances between the normal line on the running locus R2 and the front and rear features Pf2 and Pf3 may be used. In this case, similarly to the embodiment shown in FIG. 11, the first evaluation result is the distance until the normal line on the running locus R2 intersects the feature point Pf8, and the distance until the normal line on the running locus R2 intersects the feature point Pf2 is the second evaluation result.

Even if the first evaluation result and the second evaluation result are determined in this way, the accuracy of the link is similarly improved, and link information suitable for use in navigation and automatic driving can be generated.

In the above-described embodiment, it is determined whether the first evaluation result is greater than the second evaluation result by a certain value or more, that is, whether the difference between the first evaluation result and the second evaluation result is a certain value or more. The link was generated based on this, but it is not limited to this. A link may be generated based on whether the first evaluation result is greater than the second evaluation result. More specifically, if the first evaluation result>the second evaluation result, the middle feature point Pf8 is not connected, and a link connecting the front and rear feature points Pf2 and Pf3 is generated. On the other hand, if the first evaluation result≦the second evaluation result, three feature points Pf2, Pf8, and Pf3 are connected to generate a link.

Further, according to the above-described embodiment, the server 10 extracts three continuous feature points in the link information generation process, but the present invention is not limited to this. Three or more continuous feature points may be extracted.

Further, according to the above-described embodiment, the server 10 that can communicate with the in-vehicle device 20 via the network N generates the feature points and the link information, but the present invention is not limited to this. The travel locus may be stored in a memory detachable from the vehicle-mounted device 20, and a device that reads the memory removed from the vehicle-mounted device 20 may be used to generate feature points and link information. Alternatively, the vehicle-mounted device 20 may generate feature points from its own travel locus and transmit the feature points to the server 10 .

Further, according to the above-described embodiment, the generation and integration of feature points and the generation of link information are performed by one server 10, but the present invention is not limited to this. The generation and integration of feature points and the generation of link information may be performed by separate servers.

Further, in the above-described embodiment, the description has been given mainly assuming a driving path in a parking lot, but the present invention is not limited to this. A general road or the like may be used as the traveling road.

It should be noted that the present invention is not limited to the above embodiments. That is, various modifications can be made without departing from the gist of the present invention.

10 server (feature point generation device, link information generation device, acquisition unit)
12 feature point generation unit (extraction unit, first generation unit, second generation unit, generation unit, integration unit)
13 Link information generator

Claims (1)

A feature point generation device for generating feature points that are candidate nodes of a travel path based on a travel trajectory of a moving object,
an acquisition unit that acquires the travel locus;
an extraction unit that extracts a straight line component from the running locus;
a first generation unit that generates, as the characteristic point, an intersection of an extension line of the extracted straight line component and an extension line of another straight line component adjacent to the straight line component in the running trajectory; feature point generator.

Images