WO2020045427A1

WO2020045427A1 - Map generation device and map generation method

Info

Publication number: WO2020045427A1
Application number: PCT/JP2019/033513
Authority: WO
Inventors: 高行渡部
Original assignee: 株式会社デンソー
Priority date: 2018-08-31
Filing date: 2019-08-27
Publication date: 2020-03-05

Abstract

A map generation device according to the present invention is provided with a reception unit 16 that receives probe data transmitted from at least one vehicle and indicating a ground object detected to be present around the vehicle by a peripheral monitoring sensor mounted on the vehicle, and that generates or updates map data on the basis of the probe data received by the reception unit. The map generation device is further provided with: feature determination units 11, 18 that determine whether the probe data includes other probe data or sufficient information that is characteristic and that is usable for alignment with the map data; and linkage units 11, 18 that link target probe data having insufficient amount of characteristic information with probe data immediately before or after the target probe data among a series of probe data acquired through the same traveling motion of the same vehicle.

Description

Map generation device and map generation method

Cross-reference of related applications

This application is based on Japanese Patent Application No. 2018-163071 filed on Aug. 31, 2018 and Japanese Application No. 2019-138796 filed on Jul. 29, 2019, the contents of which are hereby incorporated by reference. Invite.

The present disclosure relates to a map generation device and a map generation method for generating map data using probe data generated by a plurality of vehicles.

As a method for generating map data used in an in-vehicle navigation device or an in-vehicle driving support device, for example, a technique described in Patent Document 1 is known. In the configuration described in Patent Literature 1, probe data acquired in an in-vehicle device mounted on a vehicle is transmitted from the in-vehicle device to a data center, and the data center performs a statistical method from a plurality of received probe data. It is configured to generate information such as a lane marking, for example, a white line, and sign information by using the information.

US Patent Publication 2018/0023961

When generating or updating the map data, a process of aligning the probe data with the map data is necessary. However, Patent Document 1 hardly describes the process of aligning the probe data. It is described that there is a method for performing positioning using GPS data included in the probe data. However, since conventionally known GPS data contains a large error, if GPS data is used for positioning, it may not be possible to perform proper positioning.
An object of the present disclosure is to provide a map generation device and a map generation method that can easily perform positioning of probe data.

In one embodiment of the present disclosure, a map generation device receives a probe data transmitted from at least one vehicle and indicating a feature existing around the vehicle detected by a surrounding monitoring sensor mounted on the vehicle. A map generation device that creates or updates map data based on the probe data received by the receiving unit, and that can be used for positioning the probe data with other probe data or map data. A characteristic determining unit that determines whether or not characteristic information is sufficiently included; and a target probe data having an insufficient amount of characteristic information is converted into a continuous probe data obtained by the same traveling of the same vehicle. And a connection unit for connecting to the probe data one before or one after the target probe data.

In one aspect of the present disclosure, the map generation method includes at least one map data generation or update method for generating or updating map data based on probe data indicating a feature existing around the vehicle detected by a surrounding monitoring sensor mounted on the vehicle. A map generation method performed using one processor, the method comprising receiving the probe data from at least one of the vehicles, and aligning the probe data with other probe data or map data. It is determined whether or not the characteristic information is sufficiently included, and the target probe data having an insufficient amount of the characteristic information is included in successive probe data obtained by the same traveling of the same vehicle. And joining the probe data to the probe data one before or one after the target probe data.

The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description with reference to the accompanying drawings. The drawing is
FIG. 1 illustrates the first embodiment, and is a diagram schematically illustrating an entire configuration of a system. FIG. 2 is a block diagram schematically showing a configuration of the vehicle-mounted device. FIG. 3 is a block diagram of a control unit of the in-vehicle device; FIG. 4 is a block diagram schematically showing a main configuration of the data center. FIG. 5 is a block diagram of a processing control device of the data center. FIG. 6 is a diagram (part 1) illustrating a process of aligning probe data with map data and a process of linking probe data. FIG. 7 is a flowchart illustrating an example of probe data connection control; FIG. 8 is a diagram (part 2) for explaining the alignment process between the probe data and the map data and the linking process of the probe data. FIG. 9 is a diagram (part 3) for explaining the alignment process between the probe data and the map data and the connection process of the probe data. FIG. 10 is a flowchart illustrating another example of the connection control of the probe data. FIG. 11 is a diagram illustrating a detection target landmark. FIG. 12 is a flowchart illustrating an example of probe data connection control according to the second embodiment.

(1st Embodiment)
Hereinafter, a first embodiment of the present disclosure will be described with reference to FIGS. 1 to 10. FIG. 1 schematically shows the entire configuration of a map generation system 1 according to the present embodiment. Here, the map generation system 1 includes a data center 2 that collects and analyzes probe data and generates or updates map data, and a group of a plurality of vehicles A traveling on a road. Specifically, the vehicle A group includes the entire general automobile such as a passenger car and a truck.

Each vehicle A is equipped with an in-vehicle device 3 for realizing the map generation system 1. As shown in FIG. 2, the in-vehicle apparatus 3 includes an in-vehicle camera 4, a position detection unit 5, various on-vehicle sensors 6, a map database 7, a communication unit 8, a storage unit 9, an operation unit 10, and a control unit 11. . The in-vehicle camera 4 is provided at, for example, a front portion of the vehicle A, and is configured to photograph at least a road condition ahead in the traveling direction. The in-vehicle camera 4 has a function as a periphery monitoring sensor.

The position detecting unit 5 has a function of detecting the position of the vehicle based on, for example, data received by a GNSS (Global Navigation Satellite System) receiver. The various in-vehicle sensors 6 detect speed information of the own vehicle, information of a traveling direction, that is, information of a direction, and the like. In addition, the vehicle-mounted camera 4 can be provided in front of and behind the vehicle A, and left and right. Further, as the type of the on-vehicle camera 4, a wide-angle camera can be adopted, and it is particularly desirable to adopt a twin-lens camera or more as the front camera.

The map database 7 stores, for example, map data for the whole country. For example, the map data includes road data indicating a road network and feature data indicating the position and type of a feature existing along the road. The feature data includes landmark information and lane mark information. The landmark information includes the type of landmark existing along the road, position coordinates, color, size, shape, and the like. As the types of landmarks, signs, signals, signs, poles, pedestrian crossings, road markings such as stop lines, manholes, and the like can be adopted. The lane mark information includes information indicating the position coordinates of the lane mark and which pattern of the lane mark is realized by a solid line, a dashed line, or a Bottom Dots pattern. The position of the lane mark is expressed as a group of coordinates of a point where the lane mark is formed, that is, a group of points. Note that, as another aspect, the lane mark may be represented by a polynomial expression. The lane mark information may be a polynomial-expressed line group. Further, the map data may include a traveling trajectory model. The traveling trajectory model is trajectory data generated by statistically integrating traveling trajectories of a plurality of vehicles. The traveling trajectory model is, for example, an average of traveling trajectories for each lane. The traveling trajectory model corresponds to data indicating a traveling trajectory that is a reference during automatic driving.

The communication unit 8 communicates with the data center 2 via a mobile communication network or using road-to-vehicle communication, for example. The storage unit 9 stores the camera image data taken by the on-board camera 4 with data such as the current position of the vehicle, the traveling speed, the traveling direction, the photographing date and time, and based on the camera image data. The generated probe data is stored. The storage unit 9 stores various data, programs, and the like. The details of the probe data will be described later. The operation unit 10 includes a touch panel, a switch, a display unit, and the like (not shown), and necessary operations are performed by a user of the vehicle A, for example, a driver.

The control unit 11 includes a computer and has a function of controlling the entire on-vehicle device 3. In this case, while the vehicle A is traveling, the control unit 11 constantly captures a road condition ahead by the vehicle-mounted camera 4 and stores the camera image data in the storage unit 9 together with the vehicle position data and the like. Then, the control unit 11 generates probe data having information on road signs and road division lines, that is, information on white lines of the road, etc. by performing image recognition processing on the camera image data, and stores the probe data in the storage unit 9. . The camera image data constitutes data detected by the periphery monitoring sensor 4. Further, the control unit 11 periodically, for example, once to several times a day, or every time a predetermined time elapses, for example, every one second to several seconds, or every time, for example, several hundred milliseconds Further, the communication unit 8 is configured to transmit the probe data stored in the storage unit 9 to the data center 2.

The control unit 11 includes a position information receiving unit 12, an image processing unit 13, a probe data generating unit 14, and a probe data transmitting unit 15, as shown in FIG. The position information receiving unit 12 receives the position information from the position detecting unit 5, that is, the GNSS signal, and acquires the current position information of the vehicle, that is, the own vehicle position information based on the GNSS signal. Remember. The image processing unit 13 performs image recognition processing on camera image data captured by the vehicle-mounted camera 4 and stored in the storage unit 9. The probe data generator 14 generates probe data based on the image recognition result of the image processor 13 and stores the probe data in the storage unit 9. The probe data transmission unit 15 transmits the probe data stored in the storage unit 9 to the data center 2 via the communication unit 8.

On the other hand, the data center 2 includes a communication unit 16, an input operation unit 17, a processing control unit 18, a storage unit 19, and a map database 20, as shown in FIG. The communication unit 16 receives the probe data through communication with the communication unit 8 of each vehicle A. The communication unit 16 has a function as a receiving unit. The input operation unit 17 has a keyboard, a mouse, a switch, a display unit, and the like (not shown), and is used by an operator to perform necessary input operations.

The processing control device 18 is mainly configured by a computer, and has a function of controlling the entire data center 2, that is, a function as a control unit. At the same time, the processing control device 18 executes processing for generating or updating map data. The storage unit 19 collects and stores probe data transmitted from each vehicle A. At this time, a huge amount of probe data is collected and stored, for example, from a group of general vehicles A traveling all over Japan. The map database 20 stores the generated high-precision map data.

{Circle around (2)} In the map data generation processing executed by the processing control device 18, processing for aligning probe data with map data is executed. In this case, when characteristic information such as information indicating the shape of a road lane marking and information on a road sign is not included in the probe data, it is difficult to align the probe data. Therefore, in the present embodiment, probe data that does not include the characteristic information, that is, probe data that is difficult to align, is connected to probe data that includes the characteristic information, that is, probe data that can be aligned. It is configured to perform a process of joining together, that is, a process of creating connected probe data, and convert probe data that is difficult to align into probe data that can be aligned. The connection probe data is stored in the storage unit 19. Details of the connection processing will be described later. The processing control device 18 has each function as a feature determination unit and a connection unit.

5, the processing control device 18 includes a probe data receiving unit 21, a probe data linking processing unit 22, a map data generation unit 23, and a map data distribution unit 24, as shown in FIG. The probe data receiving unit 21 receives the probe data transmitted from each vehicle A via the communication unit 16 and stores the probe data in the storage unit 19. The probe data connection processing unit 22 performs a detection process of determining whether or not the probe data stored in the storage unit 19 includes characteristic information, and converts the probe data that does not include characteristic information into characteristic information. Then, a process of connecting and linking to the probe data including is performed, and the linked probe data is stored in the storage unit 19. The map data generation unit 23 generates map data based on the probe data and the connected probe data stored in the storage unit 19, and updates the map data. The generated or updated map data is stored in the map database 20. The map data distribution unit 24 transmits the generated or updated map data to each vehicle A via the communication unit 16.

The control unit 11 of the vehicle-mounted device 3 of each vehicle A converts the camera image data captured by the vehicle-mounted camera 4 during each time the vehicle A travels a predetermined distance or every time a predetermined time elapses. A recognition process is performed to execute a process of generating probe data. In this case, the process of recognizing the camera image data and the process of generating the probe data are set to be executed, for example, every 100 milliseconds. The control unit 11 is configured to collectively transmit, for example, 400 ms of probe data to the data center 2, that is, to transmit the probe data to the data center 2 every 400 ms, for example. ing. Further, as the probe data, for example, data of a lane marking shape of a road, in this case, data including information such as a line type and a color of a lane marking, and data of a position of a road sign or a sign, Data including information such as the size, type, meaning, and content of signs and signboards, and data on the position of poles such as telephone poles, in this case, data including information such as the thickness, height, and type of poles such as telephone poles, and stopping Probe data including various data such as data of a road marking such as a line and a pedestrian crossing and data of a position of a traffic signal, in this case, data including information such as a size and a type of the traffic signal, is generated. Further, the probe data may include travel locus data indicating the travel locus of the vehicle A. That is, at least one or more of the various types of data is recognized by the image recognition processing of the camera image data, and is configured as probe data including the data of the recognition result. The movement locus corresponds to data in which the position coordinates of the own vehicle are arranged in time series. The position information of the own vehicle may be calculated based on, for example, GPS information. Of course, it may be specified by dead reckoning or localization. Here, the localization means relative position information between the landmark imaged by the on-board camera 4 and the own vehicle, for example, information indicating the distance between the landmark and the coordinates of the landmark registered in the map data. And the process of specifying the position of the own vehicle based on The relative position of the landmark imaged by the vehicle-mounted camera 4 with respect to the own vehicle can be specified by analyzing the image captured by the vehicle-mounted camera 4.

In the present embodiment, the probe data is generated based on the camera image data captured by the on-board camera 4. However, the present invention is not limited to this. For example, a millimeter wave radar or A rider, in particular, a SPAD rider or the like may be mounted, and probe data may be generated based on a detection result by the millimeter wave radar or the rider. In this case, the probe data may be generated by using both the image recognition processing of the camera image data and the detection result by the millimeter wave radar or lidar, or the probe data may be generated based only on the detection result by the millimeter wave radar or lidar. You may. Millimeter-wave radars and riders have a function as a periphery monitoring sensor. The detection results by the millimeter-wave radar and the lidar constitute data detected by the surrounding monitoring sensor 4. As the peripheral monitoring sensor, various sensors that can detect a feature around the vehicle, such as a vehicle-mounted camera, a millimeter-wave radar, and a rider, can be used. As the periphery monitoring sensor, a sensor that forms a detection range behind or on the side of the vehicle can also be used. The control unit 11 may be configured to generate probe data using a combination of a plurality of types of peripheral monitoring sensors. For example, in a configuration in which the in-vehicle camera 4 and the millimeter wave radar are used in combination as the peripheral monitoring sensor, the detection accuracy of the distance from the landmark can be improved. Further, even in an environment in which the recognition accuracy of landmarks by the in-vehicle camera 4 is deteriorated, such as at night, the recognition rate of landmarks can be secured by using the detection results of the millimeter wave radar complementarily. The same applies to a configuration in which the in-vehicle camera 4 and the rider are used together.

Here, the detection target, that is, the detection target landmark detected by the vehicle-mounted camera 4, the millimeter-wave radar, the lidar, or the like when creating the probe data will be described with reference to FIG. As shown in FIG. 11, when the detection target is a sign, it is configured to detect the shape of the sign, for example, a circle, a square, a diamond, a triangle, an octagon, and the like, and to detect the meaning of the sign if necessary. ing.

In addition, when the detection target is a signboard, the signboard is configured to detect, for example, a blue, yellow, or green signboard and, if necessary, to detect the meaning of the sign or character on the signboard. When the detection target is a road marking, for example, an arrow, a diamond, a stop, a speed, a stop line, a pedestrian crossing, a white line, or the like drawn on the road is detected. When the detection target is a signal, for example, it is configured to detect a three-lamp signal or the like. When the detection target is a group of poles, for example, it is configured to detect a telephone pole, a streetlight, and the like.

{Circle around (4)} Each probe data is added with a vehicle ID and a time stamp of the vehicle A, that is, data information that indicates the time series of the order in which the probe data is created. Note that serial number information may be added instead of the time stamp information. This makes it possible to specify and extract continuous probe data in the same traveling of the same vehicle, that is, traveling from the start to the stop of the engine, from the probe data collected in the data center 2. Have been. In the present embodiment, the connection processing is performed on the probe data of the same traveling of the same vehicle.

パターン Pattern (a) in FIG. 6 shows an example of probe data generated in one traveling of one vehicle A. The vertical lines in the pattern (a) in FIG. 6 indicate the boundaries of the probe data. In this example, for example, eight pieces of probe data P1, P2,... Is transmitted to the data center 2.

The processing control device 18 of the data center 2 performs a process of aligning the probe data P1, P2,... P8 received from the vehicle A with the map data. FIG. 6B shows eight data blocks M1, M2,... M8 of map data corresponding to the eight probe data P1, P2,. In this case, six probe data P2,... P7 excluding both ends of the eight probe data P1, P2,... P8, and six data blocks M2,. Align.

In the above-described alignment process, since the probe data P2, P3, and P7 include characteristic information, for example, data of a lane marking, specifically, data of an oblique lane marking, It can be compared with the data blocks M2, M3, M7 of the corresponding map data, that is, aligned.

On the other hand, since the probe data P4, P5, and P6 do not include data of characteristic information, specifically, the shape of the lane marking of the road is straight and any one of the probe data P4, P5, and P6 is used. Therefore, it is difficult to compare with the data blocks M4, M5, and M6 of the corresponding map data, that is, to perform alignment. In this case, the positioning can be performed by using the GPS information. However, there is a problem that the accuracy of the positioning is low. In practice, the positioning using the GPS information is difficult.

Therefore, in the present embodiment, the probe data P4, P5, and P6 that do not include the characteristic information, that is, the probe data P4, P5, and P6 that are difficult to position are converted into the characteristic information before or after the characteristic data. The probe data P3 and P7 that are included, that is, the probe data P3 and P7 that can be aligned, are connected and connected. For example, the probe data P4 and the preceding probe data P3 including characteristic information are connected to create the connected probe data P3-4. It is also preferable that the probe data P5, P4 and the probe data P3 are connected to create the connected probe data P3-4-5. Further, the probe data P6, P5, and P4 may be joined to the probe data P3 to create the connected probe data P3-4-5-6.

{Circle around (4)} The probe data P6 may be connected to the probe data P7 including characteristic information thereafter to create the connected probe data P6-7. Further, the probe data P5, P6 and the probe data P7 may be joined to create the connected probe data P5-6-7. Further, the probe data P4, P5, and P6 may be joined with the probe data P7 to create the connected probe data P4-5-6-7.

FIG. 7 is a flowchart showing the control contents of an example of the probe data linking process. The flowchart of FIG. 7 shows the contents of control by the processing control device 18 of the data center 2. First, in step S10 of FIG. 7, the processing control device 18 determines whether characteristic information, for example, information on the shape of a road division line, is included in the probe data. Is performed.

Then, the process proceeds to step S20, and it is determined whether or not characteristic information is included in the probe data. Here, when characteristic information is not included (NO), the process proceeds to step S30, and one of the probe data of the same traveling of the same vehicle with respect to the probe data currently being processed, that is, the target probe data. The previous probe data or the next probe data is linked.

Subsequently, the process returns to step S10 to determine whether characteristic information, for example, information on the shape of a lane marking, is included in the connected probe data, that is, a process of detecting the characteristic information from the connected probe data. Do. Then, the above-described processing is repeatedly executed. That is, the connection processing of the probe data is continued until the characteristic information is included in the connection probe data. If the probe data including the characteristic information has been identified, in step S30, the probe data including the characteristic information is collectively linked to the target probe data. You may comprise.

In step S20, when characteristic information is included in the probe data or the linked probe data (YES), the process proceeds to step S40 without executing the probe data linking process, and the next probe data is It is determined whether or not there is. Here, when there is the next probe data (YES), the process returns to step S10, and the above-described processing is repeatedly performed on the next probe data. In step S40, when there is no next probe data, the process proceeds to “NO” and the present control ends.

FIG. 8 shows another example of the probe data linking process, for example, a case in which data of a lane marking of a road is used as characteristic information. FIG. 8A shows an example of probe data generated in one traveling of one vehicle A. In this example, for example, eight pieces of probe data P11, P12,... P18 are generated, and these probe data are transmitted to the data center 2.

Then, the processing control device 18 of the data center 2 performs a process of aligning the probe data P11, P12,... P18 received from the vehicle A with the map data. The pattern (b) of FIG. 8 shows eight data blocks M11, M12,... M18 of map data corresponding to the eight probe data P11, P12,. In this case, as shown in FIG. 8, six probe data P12,... P17 excluding both ends of the eight probe data P11, P12,. Blocks M12,... M17 are aligned.

In the above-described alignment process, the probe data P12 and P13 include characteristic information, for example, data of a lane marking shape of a road, specifically, data of a lane marking lane shape. The data can be compared with the data blocks M12 and M13, that is, the data can be aligned.

Since the probe data P14 and P15 do not include data of characteristic information, specifically, since the lane markings of the road are straight and indistinguishable, the data blocks M14 and M15 of the corresponding map data are not included. It is difficult to compare, that is, align.

Here, the probe data P16 and P17 include information characteristic of the lane marking of the road. Specifically, it includes data of road markings for which traffic is prohibited. However, this characteristic information does not exist in the data blocks M16 and M17 of the corresponding map data. That is, the road marking data is new data that does not exist in the current map data, and is data to be updated. Therefore, it is difficult to compare the probe data P16 and P17 with the data blocks M16 and M17 of the map data alone, that is, to align the probe data P16 and P17. In this case, the characteristic information of the probe data P16 and P17 is not characteristic information that can be used for alignment, but the probe data includes characteristic information that can be used for alignment with map data. Can not be determined to be sufficiently contained. The state where the characteristic information is not sufficiently included indicates a case where the characteristic information is less than three, such as one or two. In addition, the case where the characteristic information is not included, that is, the case where the characteristic information is zero is also included.

In the present embodiment, the probe data P14 and P15 that do not include characteristic information and are difficult to position, and the probe data that includes characteristic information but are difficult to position, that is, the amount of characteristic information Insufficient probe data P16 and P17 are connected and linked to probe data including characteristic information that can be used before or after alignment, that is, probe data P13 that can be aligned.

For example, the probe data P14, P15, P16, and P17 are all connected and connected, and furthermore, a probe capable of performing positioning including characteristic information before the four connected probe data P14, P15, P16, and P17. The four probe data P14, P15, P16, and P17 are connected and connected to the data P13 to create probe data P13-14-15-16-17. When there is probe data that can be aligned after the four probe data P14, P15, P16, and P17, the probe data after the alignment is added to the four probe data P14, P15, P16 and P17 may be connected and connected.

FIG. 9 shows another example of the probe data linking process, for example, a case where road sign data is used as characteristic information. FIG. 9A shows an example of probe data generated in one traveling of one vehicle A. In this example, for example, eight pieces of probe data P21, P22,... P28 are generated, and these probe data are transmitted to the data center 2.

Then, the processing control device 18 of the data center 2 performs a process of aligning the probe data P21, P22,... P28 received from the vehicle A with the map data. The pattern (b) in FIG. 9 shows eight data blocks M21, M22,... M28 of map data corresponding to the eight probe data P21, P22,. In this case, as shown in FIG. 9, six probe data P22,... P27 excluding both ends of the eight probe data P21, P22,. Blocks M22,... M27 are aligned.

In the alignment processing, the probe data P22, P23, and P27 are characteristic information, for example, data of road signs, specifically, data indicating the type, position, size, and number of road signs are compared. It matches the data of the road sign included in the data blocks M22, M23, and M27 of the map data. Therefore, both can be compared, that is, alignment can be performed. In the case of this configuration, the characteristic information included in the probe data P22, P23, and P27 is characteristic information that can be used for alignment.

On the other hand, the probe data P24, P25, and P26 include road sign data as characteristic information. However, this characteristic information, that is, the data of the road sign included in the probe data P24, P25, and P26 is the data of the road sign included in the data blocks M24, M25, and M26 of the corresponding map data and the shape of the road division line. Does not match the data in. Specifically, the position and number of road signs, the shape of the road marking line, and the like do not match. Therefore, it is difficult to compare the probe data P24, P25, and P26 with the data blocks M24, M25, and M26 of the map data alone, that is, to align the probe data. That is, the characteristic information of the probe data P24, P25, and P26 is insufficient as characteristic information that can be used for alignment. That is, the probe data P24, P25, and P26 do not sufficiently include characteristic information that can be used for alignment with the map data.

Therefore, in the present embodiment, the probe data P24, P25, and P26 that include characteristic information but are difficult to align, that is, probe data with an insufficient amount of characteristic information, The probe data including characteristic information usable for alignment, that is, probe data P23 or P27 that can be aligned is connected and connected.

For example, the probe data P24, P25, and P26 are all connected and connected, and further, the probe data P23 that includes the characteristic information before the three connected probe data P24, P25, and P26 and can be aligned, The three probe data P24, P25, and P26 are connected and connected to create probe data P23-24-25-26.

Also, in the present embodiment, the probe data P27 that can be positioned exists after the three pieces of probe data P24, P25, and P26 that have been connected. Therefore, the probe data P24, P25, and P26 may be connected and connected to the probe data P27 that can be aligned to create probe data P24-25-26-27.

(4) The first probe data P24 among the three probe data P24, P25, and P26 is connected to the preceding probe data P23 that includes characteristic information and can be aligned. Then, the last two pieces of probe data P25 and P26 are connected, and further, the two pieces of connected probe data P25 and P26 are connected to probe data P27 that includes the characteristic information thereafter and can be aligned. You may make it connect.

In addition, the previous two probe data P24 and P25 among the three probe data P24, P25 and P26 are connected, and the two connected probe data P24 and P25 are distinguished by a characteristic The probe data P23 containing information and capable of alignment is connected and connected. Then, the last one piece of probe data P26 may be connected to and connected to probe data P27 which includes the subsequent characteristic information and can be aligned.

FIG. 10 is a flowchart showing the control contents of an example of the probe data linking process of FIG. The flowchart of FIG. 10 shows the contents of control by the processing control device 18 of the data center 2. First, in step S110 of FIG. 10, the processing control device 18 performs a process of detecting whether or not characteristic information is included in the probe data, that is, a process of detecting the characteristic information from the probe data. As characteristic information, for example, as shown in FIG. 11, data on the shape of a lane marking, data on the line type and color of the lane marking, data on the manner of assigning lane markings, and branching and merging indicating the road shape , Curves, etc., data of road sign layout mode, data of road sign size, type, content, etc., data of signboard layout mode, data of signboard size, type, content, etc., road marking application mode Data etc. are included. The sign may be a direction sign or a commercial advertising sign. It is preferable that the signboard information includes character string information such as numbers and alphabets. Note that the characteristic information includes characteristic information that cannot be aligned, that is, characteristic information that cannot be used for alignment.

Then, the process proceeds to step S120, where it is determined whether characteristic information is included in the probe data. Here, when characteristic information is not included (NO), the process proceeds to step S130, where the probe data immediately before the probe data of the same traveling of the same vehicle with respect to the probe data currently being processed, Alternatively, the next probe data is connected and connected.

Then, the process returns to step S110 to perform a process of detecting whether or not the above-described characteristic information is included in the connected probe data, that is, a process of detecting the above-described characteristic information from the connected probe data. Then, the above-described processing is repeatedly executed. That is, the connection processing of the probe data is continued until the characteristic information is included in the connection probe data. If the probe data including the characteristic information has been identified, in step S130, the probe data including the characteristic information is collectively connected to the target probe data. You may comprise.

{Circle around (5)} In step S120, when characteristic information is included in the probe data or the connected probe data (YES), the process proceeds to step S150. In this step S150, the processing control device 18 determines the characteristic information included in the probe data, for example, the data of the road sign, and the characteristic information included in the data block of the corresponding map data, for example, the data of the road sign. To see if they match. In this case, the characteristic information included in the probe data and the characteristic information included in the corresponding data block of the map data may be configured to detect the number of matching characteristic information. .

Then, the process proceeds to step S160, and it is determined whether the characteristic information included in the probe data matches the characteristic information included in the data block of the corresponding map data. In this case, the number of pieces of characteristic information matching between the characteristic information included in the probe data and the characteristic information included in the data block of the corresponding map data is specified, and the matching characteristic is determined. It is determined whether or not the number of pieces of target information is equal to or more than the number of pieces of data that can three-dimensionally specify the data block of the probe data, that is, map data, for example, three or more. In this embodiment, it is configured to determine whether or not there are three or more pieces of matching characteristic information. However, the present invention is not limited to this. Alternatively, it may be configured to determine whether there are three points, whether there are four points, whether there are five points, or whether there are six or more points. That is, in step S160, it is configured to determine whether or not the probe data sufficiently includes characteristic information that can be used for alignment with the map data.

In step S160, when the characteristic information does not match between the probe data and the data block of the corresponding map data, that is, when there is no matching characteristic information of three or more points, in other words, the probe currently being processed is When the amount of characteristic information included in the data is insufficient (NO), the process proceeds to step S170. In this step S170, the probe data that is currently being processed is connected to the previous probe data or the next probe data of the same vehicle probe data of the same traveling.

Then, returning to step S150, the characteristic information included in the connected probe data, for example, the data of the road sign, is compared with the characteristic information included in the data block of the corresponding map data, for example, the data of the road sign. To see if they match. Then, the processing control device 18 repeatedly executes the above-described processing. This allows the probe data connection process to be continued until the characteristic information included in the connected probe data matches the characteristic information included in the corresponding data block of the map data. ing. If the probe data including the characteristic information matching the characteristic information of the map data has been identified, in step S170, up to the probe data including the characteristic information, It may be configured to connect to the probe data to be performed.

Also, in the above step S160, when the characteristic information matches between the probe data or the connected probe data and the data block of the corresponding map data, that is, when there are three or more matching characteristic information (YES), Proceeding to step S180, it is determined whether or not there is next probe data without performing the linking process. Here, when there is the next probe data (YES), the process returns to step S110, and the above-described processing is repeatedly performed on the next probe data. In step S180, when there is no next probe data, the process proceeds to “NO” and the present control ends.

According to the present embodiment having the above-described configuration, the processing control device 18 detects whether characteristic information is included in the probe data, and converts the target probe data that does not include the characteristic information to the same vehicle. The probe data is connected to the probe data immediately before or after the target probe data in the continuous probe data obtained by the same traveling. According to this configuration, the probe data can be easily and reliably aligned by using the connected probe data instead of the probe data that cannot be aligned.

According to the present embodiment, as the characteristic information, characteristic information usable for aligning the probe data with the map data is used. With this configuration, the probe data linking process as shown in FIG. 8 can be realized, so that the alignment of the probe data can be performed more reliably.

Further, according to the present embodiment, the processing control device 18 matches the characteristic information included in the probe data with the characteristic information included in the data block of the corresponding map data. Is configured to detect whether the number of pieces of information is equal to or more than the number of pieces of data that can specify the probe data three-dimensionally, that is, the probe data contains sufficient characteristic information that can be used for alignment with the map data. It is configured to determine whether or not it is included. Furthermore, according to the present embodiment, target probe data in which matching characteristic information does not exceed the number of data that can be specified three-dimensionally, that is, target probe data with an insufficient amount of characteristic information, The probe data is connected to the probe data immediately before or after the target probe data in the continuous probe data obtained by the same traveling. According to this configuration, the probe data linking process as shown in FIG. 9 can be realized, so that the alignment of the probe data can be performed more reliably.

Further, according to the present embodiment, the probe data linking process is configured to be executed by the processing control device 18 of the data center 2. Instead, the probe data linking process is performed by the in-vehicle device 3 of each vehicle A. May be configured to be executed by the control unit 11. That is, the control unit 11 of the in-vehicle device 3 may be configured to have each function as a feature determination unit and a connection unit. When configured in this manner, the in-vehicle device 3 of each vehicle A is configured to transmit the probe data and the connected probe data to the data center 2.

(2nd Embodiment)
FIG. 12 shows a second embodiment. The same components as those in the first embodiment are denoted by the same reference numerals. In the first embodiment, as shown in FIGS. 6, 7, 8, 9 and 10, the probe data P1, P2,... P8 received from one vehicle A are aligned with the map data. However, the present invention is not limited to this. In the second embodiment, probe data P1, P2,... P8 received from one vehicle A and probe data received from another vehicle A, that is, probe data P1, P2 of another probe data or another trip. ,..., P8.

Specifically, in the second embodiment, eight data blocks of the map data shown in the pattern (b) of FIG. 6, the pattern (b) of FIG. 8, and the pattern (b) of FIG. M1, M2,... M8, eight data blocks M11, M12,... M18, eight data blocks M21, M22,. , P80, eight probe data P110, P120,... P180 of another trip, and eight probe data P210, P220,. What is necessary is just to perform the process which performs.

FIG. 12 replaces the eight data blocks M21, M22,... M28 of the map data shown in the pattern (b) in FIG. 9 with eight probe data P210, P220,. 9 is a flowchart illustrating control contents of a probe data linking process in the alignment process. The flowchart of FIG. 12 shows the contents of control by the processing control device 18 of the data center 2. First, the processing from step S110 to step S130 in FIG. 12 is executed in the same manner as the processing from step S110 to step S130 in FIG. 10 of the first embodiment.

Thereafter, when the process proceeds to step S250, the processing control device 18 determines the characteristic information included in the probe data received from one vehicle A, for example, the data of the road sign and the corresponding probe information received from the other vehicle A. Data, that is, characteristic information included in the probe data of another trip, for example, data of a road sign, is compared to determine whether or not the two match. In this case, the characteristic information included in the probe data received from one vehicle A and the characteristic information included in the corresponding probe data received from another vehicle A correspond to the characteristic information. May be configured to detect the number. Hereinafter, the probe data received from one vehicle A is referred to as “one probe data”, and the probe data received from another vehicle A is referred to as “other trip probe data”.

Then, the process proceeds to step S260, and it is determined whether or not the characteristic information included in one probe data and the characteristic information included in the corresponding probe data of another trip match. In this case, the number of pieces of characteristic information that match between the characteristic information included in one probe data and the characteristic information included in the corresponding probe data of another trip is specified. It may be configured to determine whether the number of matching characteristic information is equal to or more than the number of data that can three-dimensionally specify one probe data and the probe data of another trip, for example, three points or more. In this embodiment, it is configured to determine whether or not there are three or more pieces of matching characteristic information. However, the present invention is not limited to this. Alternatively, it may be configured to determine whether there are three points, whether there are four points, whether there are five points, or whether there are six or more points. That is, in step S260, it is configured to determine whether or not one probe data sufficiently contains characteristic information that can be used for alignment with probe data of another trip.

In step S260, when characteristic information does not match between one probe data and the corresponding probe data of another trip, that is, when there is no matching characteristic information of three or more points, in other words, if the current processing is being performed. If the amount of characteristic information included in the probe data is insufficient (NO), the process proceeds to step S270. In this step S270, the previous probe data or the next probe data of the same traveling probe data of the same vehicle is connected to the currently processed probe data and connected. Further, in this step S270, the probe data of the other trip immediately before the probe data of the other trip of the same traveling of the same vehicle with respect to the probe data of the other trip that is currently being processed, or the probe data of the next trip And connect the probe data of other trips.

Then, returning to step S250, the characteristic information included in the connection probe data, for example, the data of the road sign, and the characteristic information included in the connection probe data of the corresponding other trip, for example, the data of the road sign, are obtained. Compare to see if they match. Then, the processing control device 18 repeatedly executes the above-described processing. Thereby, until the characteristic information included in the connected probe data matches the characteristic information included in the corresponding connected probe data of another trip, the probe data received from one vehicle A becomes identical. The connection processing and the connection processing of the probe data of another trip received from another vehicle A are continued. If the probe data including the characteristic information that matches the characteristic information of the probe data of another trip has been identified, in step S270, the probe data including the characteristic information is collectively collected. It may be configured to be linked to the target probe data. In this case, similarly, it is preferable that the probe data of another trip including the characteristic information be collectively connected to the probe data of the target other trip.

In step S260, when the characteristic information of one probe data or one connected probe data and the corresponding probe data of another trip match, that is, when there are three or more matching characteristic information, (YES), the process proceeds to step S180 to determine whether or not there is the next probe data without performing the connection process. Here, when there is the next probe data (YES), the process returns to step S110, and the above-described processing is repeatedly performed on the next probe data.

{Circle around (2)} If there is no next probe data in step S180 (NO), the flow proceeds to step S290. In this step S290, probe data for which connection processing has been performed on the same road segment as described above is prepared for a plurality of vehicles, and statistical processing is performed on the probe data for which connection processing has been performed for these vehicles. Execute In this statistical processing, for example, a normal averaging processing may be executed, or the statistical processing described in Japanese Patent Application No. 2018-163076 previously filed by the present applicant may be executed. good. Note that probe data created by performing statistical processing on probe data that has undergone connection processing for a plurality of vehicles is referred to as integrated probe data or statistical probe data. The road segment is a road management unit in the map data. The road segment is obtained by dividing a road according to a predetermined rule. The road segment may be a road segmented by a predetermined length, for example, every 10 m.

Note that the processing control device 18 executing step S290 identifies a plurality of the probe data for the same road segment using the probe data connected so that other probe data can be aligned, and This corresponds to an integrated processing unit that generates integrated probe data by statistically processing a plurality of probe data. The statistical process corresponds to a process of averaging the position coordinates of each feature included in each probe data or obtaining a median of the position coordinates of each feature. In addition, the statistical processing of the probe data may include a process of generating a traveling trajectory model by averaging traveling trajectories included in each probe data.

Thereafter, the process proceeds to step S300, where characteristic information included in the integrated probe data, for example, data of a road sign, and characteristic information included in a data block of the corresponding map data, for example, data of a road sign Then, it is determined whether or not the two coincide with each other, and a connection process of the integrated probe data is executed. The integrated probe data linking process is configured to be executed in substantially the same manner as the processes from step S150 to step S180 in FIG. In this case, the process of replacing the eight pieces of probe data received from one vehicle A shown in the pattern (a) in FIG. 9 with the integrated probe data statistically processed in the above-described step S290 is performed. It is configured. After execution of the process of step S300, the present control ends.

Note that, as a subsequent process, the processing control device 18 detects a difference between the current map data and the integrated probe data by using the result of the alignment between the integrated probe data and the map data. Hereinafter, this different portion is referred to as a “map change point”. The map change point indicates a portion that may be different between the current map data and the real world. For example, a map change point indicates that a feature that has been relocated, newly constructed, or removed exists. When there is a map change point, the processing control device 18 generates, for example, map data in which the current map data reflects the map change point. According to such a configuration, the processing control device 18 can update the map data based on the probe data sequentially uploaded from a plurality of vehicles. Further, since the integrated probe data is obtained by integrating a plurality of probe data, it has more statistical certainty than individual probe data. Therefore, according to the configuration in which the map data is updated using the integrated probe data, the possibility of changing the map data to erroneous contents can be reduced as compared with the configuration in which the map data is updated using one probe data. Further, the processing control device 18 can generate and update the traveling trajectory model by connecting and statistically processing a plurality of probe data. That is, when a change occurs in a feature or road shape existing along a road, a traveling trajectory model corresponding to the change can be quickly generated and distributed. The travel trajectories included in the probe data are connected together with the connection processing of the probe data, and the travel trajectories included in the respective connected probe data are integrated by statistical processing to form a travel trajectory model. .

構成 In addition, the configuration of the second embodiment other than the above is the same as the configuration of the first embodiment. Therefore, also in the second embodiment, substantially the same operation and effect as in the first embodiment can be obtained. In particular, in the second embodiment, one probe data for a plurality of vehicles received from a plurality of vehicles is connected, and further, integrated probe data obtained by statistically processing the plurality of probe data is combined with map data. It is configured to perform the alignment process and execute the linking process of the integrated probe data. Therefore, the positioning and linking of the integrated probe data can be easily and reliably performed.

Further, in each of the above embodiments, as shown in FIGS. 7, 10, and 12, the probe data received from one vehicle A is configured to execute the connection processing of the probe data, but is not limited thereto. Instead, the same linking process may be performed on integrated probe data created by statistical processing. In this case, in each of the flowcharts of FIG. 7, FIG. 10, and FIG. 12, it is preferable to perform control so that the probe data received from one vehicle A is replaced with integrated probe data to execute each process.

Although the present disclosure has been described based on the embodiments, it is understood that the present disclosure is not limited to the embodiments and the structure. The present disclosure also encompasses various modifications and variations within an equivalent range. In addition, various combinations and forms, and other combinations and forms including only one element, more or less, are also included in the scope and spirit of the present disclosure.

The control unit and the technique according to the present disclosure are realized by a dedicated computer provided by configuring a processor and a memory programmed to execute one or a plurality of functions embodied by a computer program. May be. Alternatively, the control unit and the technique described in the present disclosure may be implemented by a dedicated computer provided by configuring a processor with one or more dedicated hardware logic circuits. Alternatively, the control unit and the method according to the present disclosure may be implemented by a combination of a processor and a memory programmed to perform one or more functions and a processor configured with one or more hardware logic circuits. It may be realized by one or more dedicated computers configured. Further, the computer program may be stored in a computer-readable non-transitional tangible recording medium as instructions to be executed by a computer.

Claims

A receiving unit (16) for receiving probe data transmitted from at least one vehicle and indicating a feature existing around the vehicle detected by a surrounding monitoring sensor mounted on the vehicle, and received by the receiving unit; A map generation device that creates or updates map data based on the probe data obtained,
A characteristic determining unit (11, 18) for determining whether the probe data sufficiently contains characteristic information usable for alignment with other probe data or map data,
Concatenation that joins target probe data having an insufficient amount of characteristic information to probe data immediately before or after the target probe data in continuous probe data obtained by the same traveling of the same vehicle. And a map generation device comprising:
2. The map generation device according to claim 1, wherein the map generation device is configured to sequentially generate the probe data based on data provided from at least one of a vehicle-mounted camera, a millimeter-wave radar, and a rider serving as the periphery monitoring sensor. 3. .
The characteristic determining unit determines that the characteristic information matching the characteristic information included in the probe data and the characteristic information included in the corresponding data block of the map data three-dimensionally converts the probe data. Is configured to detect whether or not the number of identifiable data is greater than or equal to,
The connection unit may be configured to convert the target probe data in which the matching characteristic information does not exceed the three-dimensionally identifiable data number into the target probe data in continuous probe data obtained by the same traveling of the same vehicle. The map generating apparatus according to claim 1, wherein the map generating apparatus is configured to be connected to probe data immediately before or after one of the above.
Using the probe data connected so that the other probe data can be aligned by the connection unit, a plurality of the probe data for the same road segment is specified, and based on the plurality of the probe data, An integration processing unit (S290) for generating integrated probe data,
4. The map generation device according to claim 1, further comprising: a map generation unit configured to generate or update map data based on the integrated probe data generated by the integration processing unit. 5.
5. The map generation device according to claim 4, wherein the map generation unit is configured to detect a map change point by comparing the integrated probe data with current map data.
A map generation method executed using at least one processor for creating or updating map data based on probe data indicating a feature existing around a vehicle detected by a surrounding monitoring sensor mounted on the vehicle. So,
Receiving the probe data from at least one of the vehicles;
To determine whether the probe data contains sufficient characteristic information that can be used for alignment with other probe data or map data,
Splicing the target probe data having an insufficient amount of characteristic information to the probe data immediately before or after the target probe data in the continuous probe data obtained by the same traveling of the same vehicle. And a map generation method including: