Map update device, map update method, and computer program for updating map

Abstract

To provide a map updating device capable of properly updating map information using data representing collected geographic features.SOLUTION: A map updating device includes: an association unit 42 that identifies one or more registered geographic features within a predetermined distance from the location of a geographic feature represented in geographic feature data received from a vehicle 2 via a communication unit 31 among individual registered geographic features represented in map information, refers to geographic feature group information representing two or more types that are similar to each other as the same group among multiple types of geographic features, and associates the geographic feature represented in the geographic feature data with a registered geographic feature of a type that is the same as, or belongs to the same group as the type of the geographic feature represented in the geographic feature data among one or more identified registered geographic features; and an updating unit 43 that updates at least one of the location or type of the registered geographic feature on the basis of the location or type of the geographic feature represented in a predetermined number of pieces of geographic feature data associated with any of the registered geographic features.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a map updating device, a map updating method, and a computer program for updating a map, which update information on features shown on a map. [Background technology]

[0002] A highly accurate map that an autonomous driving system for a vehicle refers to in order to control the autonomous driving of the vehicle is required to accurately represent information about roads or features around the roads that are related to the vehicle's travel. Therefore, a technology has been proposed to reliably and accurately update the map when changes occur (see Patent Document 1).

[0003] In the technology disclosed in Patent Document 1, a server device stores an advanced map DB including feature information on features in a storage unit. The server device receives difference information indicating a difference between the feature information and the actual feature corresponding to the feature information from a plurality of vehicle-mounted devices equipped with external sensors that measure the features. The server device also transmits a raw data request signal to the vehicle-mounted devices, requesting transmission of raw data, which is measurement data of the actual feature, according to the reliability calculated based on the plurality of difference information. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-73547 Summary of the Invention [Problem to be solved by the invention]

[0005] Some types of features are similar to other types of features. For such features, the type of feature may be erroneously recognized when data representing the feature is generated on the vehicle side, or the type of feature may be erroneously recorded on the map before the update. In such a case, if the features represented in the collected data are matched only with the same type of features recorded on the map before the update, the matching may not be performed even if there is a feature on the map that should be matched. As a result, the map may not be updated accurately, or the amount of data not used for updating the map may increase.

[0006] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to provide a map updating device capable of appropriately updating map information using collected data representing features. [Means for solving the problem]

[0007] According to one embodiment, there is provided a map updating device, which includes: a storage unit that stores the positions of each registered feature represented in map information, and feature group information that represents two or more similar features among a plurality of feature types as the same group; a correspondence unit that identifies one or more registered features that are located within a predetermined distance from the positions of the features represented in feature data received from the vehicle via a communication unit among the individual registered features represented in the map information, and refers to the feature group information to correspond the features represented in the feature data to registered features of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature; and an update unit that updates at least one of the position or type of at least one of the individual registered features represented in the map information, based on the position or type of the feature represented in a predetermined number of feature data associated with the registered feature.

[0008] In this map updating device, the feature group information represents two or more levels of groups with different degrees of similarity between the types of features belonging to them, and it is preferable that the matching unit searches for and matches registered features of a type that belongs to the same group as the type of feature represented in the feature data, from among the one or more identified registered features, in order of group with the highest degree of similarity.

[0009] Furthermore, in this map updating device, the feature group information represents two or more levels of groups with different degrees of similarity between the types belonging to the group, and it is preferable that the update unit increases the weight in updating the position or type of at least any of the individual registered features represented in the map information, the higher the degree of similarity of the group that is the same as the type of the registered feature to the features represented in each of a predetermined number of feature data associated with that registered feature.

[0010] In addition, in this map updating device, it is preferable that the feature data further has a confidence level for each of two or more types of features represented in the feature data, which indicates the likelihood that the feature is of that type, and the matching unit searches for and matches registered features of a type that belongs to the same group as the type represented in the feature data, in order of highest confidence level for the feature.

[0011] According to another aspect, there is provided a map updating method, which includes: identifying, among the individual registered features represented in map information, one or more registered features located within a predetermined distance from a position of a feature represented in feature data received from a vehicle via a communication unit, referring to feature group information that represents two or more similar types of features as the same group among a plurality of types of features, and associating the feature represented in the feature data with a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature, among the one or more identified registered features, and updating, for at least any of the individual registered features represented in the map information, at least one of the position or the type of the registered feature based on the position or the type of the feature represented in a predetermined number of feature data associated with the registered feature.

[0012] According to yet another aspect, there is provided a computer program for updating a map, the computer program for updating the map including instructions for causing a computer to execute the following: identify, among the individual registered features represented in the map information, one or more registered features located within a predetermined distance from a position of a feature represented in feature data received from a vehicle via a communication unit, refer to feature group information that represents two or more similar types of features as the same group among a plurality of types of features, and associate the feature represented in the feature data with a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature, among the one or more identified registered features; and update, for at least any of the individual registered features represented in the map information, at least one of the position or the type of the registered feature based on the position or the type of the feature represented in a predetermined number of feature data associated with the registered feature. Effect of the Invention

[0013] The map updating device according to the present disclosure provides an effect of being able to appropriately update map information using collected data representing features. [Brief description of the drawings]

[0014] [Figure 1] 1 is a schematic configuration diagram of a feature data collection system in which a map updating device is implemented. [Diagram 2] FIG. 1 is a schematic configuration diagram of a vehicle. [Diagram 3] FIG. 2 is a hardware configuration diagram of a data acquisition device. [Figure 4] FIG. 2 is a hardware configuration diagram of a server that is an example of a map updating device. [Diagram 5] FIG. 4 is a functional block diagram of a processor of a server for a map update process. [Figure 6] FIG. 11 is a diagram showing an example of feature group information. [Figure 7] 13(a) to 13(c) are diagrams showing examples of associations between features represented in collected feature data and registered features. [Figure 8]13 is an operation flowchart of a map update process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, a map updating device, a map updating method executed by the map updating device, and a computer program for updating the map will be described with reference to the drawings. The map updating device collects data (hereinafter referred to as feature data) representing the positions and types of features related to the travel of the vehicles from one or more vehicles capable of communication for a predetermined area. The map updating device then updates the positions or types of features represented in the map information based on the collected feature data. Hereinafter, for convenience of explanation, the features represented in the map information may be referred to as registered features. The map updating device identifies registered features located within a predetermined distance from the positions of the features represented in the received feature data among the individual registered features represented in the map information to be updated. Furthermore, the map updating device refers to feature group information and associates the features represented in the received feature data with registered features of the same type as the features represented in the received feature data or registered features of the same type as the features represented in the received feature data among the identified registered features. The map updating device then updates at least one of the position and type of each registered feature represented in the map information, based on features represented in a predetermined number or more of feature data associated with the registered feature.

[0016] The features related to vehicle travel that are the detection targets include, for example, various road signs, various road markings, traffic lights, and other features related to vehicle travel. The feature group information is information that represents two or more similar types of features as the same group.

[0017] FIG. 1 is a schematic diagram of a feature data collection system in which a map updating device is implemented. In this embodiment, the feature data collection system 1 has at least one vehicle 2 and a server 3 which is an example of a map updating device. Each vehicle 2 is connected to the server 3 via the wireless base station 5 and the communication network 4 by accessing, for example, a wireless base station 5 connected to the communication network 4 to which the server 3 is connected via a gateway (not shown). Note that, for simplification, only one vehicle 2 is illustrated in FIG. 1, but the feature data collection system 1 may have a plurality of vehicles 2. Similarly, only one wireless base station 5 is illustrated in FIG. 1, but a plurality of wireless base stations 5 may be connected to the communication network 4. In addition, the server 3 may be communicably connected to a traffic information server (not shown) which manages traffic information via the communication network.

[0018] 2 is a schematic diagram of the vehicle 2. The vehicle 2 has a camera 11, a GPS receiver 12, a wireless communication terminal 13, and a data acquisition device 14. The camera 11, the GPS receiver 12, the wireless communication terminal 13, and the data acquisition device 14 are communicatively connected via an in-vehicle network conforming to a standard such as a controller area network. The vehicle 2 may further have a navigation device (not shown) that searches for a planned travel route of the vehicle 2 and navigates the vehicle 2 so that the vehicle 2 travels along the planned travel route.

[0019] The camera 11 is an example of an imaging unit for capturing images of the surroundings of the vehicle 2, and includes a two-dimensional detector configured with an array of photoelectric conversion elements sensitive to visible light, such as a CCD or C-MOS, and an imaging optical system for forming an image of the area to be captured on the two-dimensional detector. The camera 11 is attached, for example, inside the cabin of the vehicle 2 so as to face the front of the vehicle 2. The camera 11 captures an image of the area in front of the vehicle 2 at a predetermined capture period (for example, 1 / 30 to 1 / 10 seconds) and generates an image showing the area in front of the vehicle 2. The image obtained by the camera 11 may be a color image or a gray image. Note that the vehicle 2 may be provided with a plurality of cameras 11 with different capture directions or focal lengths.

[0020] Each time the camera 11 generates an image, it outputs the generated image to the data acquisition device 14 via the in-vehicle network.

[0021] The GPS receiver 12 receives GPS signals from GPS satellites at predetermined intervals and determines the own position of the vehicle 2 based on the received GPS signals. Then, the GPS receiver 12 outputs positioning information indicating the positioning result of the own position of the vehicle 2 based on the GPS signals to the data acquisition device 14 via the in-vehicle network at predetermined intervals. Note that the vehicle 2 may have a receiver that complies with a satellite positioning system other than the GPS receiver 12. In this case, it is sufficient that the receiver determines the own position of the vehicle 2.

[0022] The wireless communication terminal 13 is an example of a communication unit, and is a device that executes wireless communication processing conforming to a predetermined wireless communication standard. For example, by accessing the wireless base station 5, the wireless communication terminal 13 is connected to the server 3 via the wireless base station 5 and the communication network 4. The wireless communication terminal 13 generates an uplink wireless signal including feature data and the like received from the data acquisition device 14. The wireless communication terminal 13 transmits the uplink wireless signal to the wireless base station 5 to transmit the feature data, driving information, and the like to the server 3. The wireless communication terminal 13 also receives a downlink wireless signal from the wireless base station 5, and passes a collection instruction or a collection stop instruction from the server 3, which are included in the wireless signal, to the data acquisition device 14 or an electronic control unit (ECU, not shown) that controls the driving of the vehicle 2.

[0023] 3 is a hardware configuration diagram of the data acquisition device. The data acquisition device 14 generates feature data based on the image generated by the camera 11. Furthermore, the data acquisition device 14 generates driving information of the vehicle 2. To this end, the data acquisition device 14 has a communication interface 21, a memory 22, and a processor 23.

[0024] The communication interface 21 is an example of an in-vehicle communication unit, and has an interface circuit for connecting the data acquisition device 14 to the in-vehicle network. That is, the communication interface 21 is connected to the camera 11, the GPS receiver 12, and the wireless communication terminal 13 via the in-vehicle network. Then, the communication interface 21 passes the received image to the processor 23 every time it receives an image from the camera 11. Also, the communication interface 21 passes the received positioning information to the processor 23 every time it receives positioning information from the GPS receiver 12. Furthermore, the communication interface 21 passes to the processor 23 a collection instruction and a collection stop instruction of feature data received from the server 3 via the wireless communication terminal 13. Furthermore, the communication interface 21 outputs the feature data received from the processor 23 to the wireless communication terminal 13 via the in-vehicle network.

[0025] The memory 22 includes, for example, a volatile semiconductor memory and a non-volatile semiconductor memory. The memory 22 may further include other storage devices such as a hard disk device. The memory 22 stores various data used in the process related to the feature data generation executed by the processor 23 of the data acquisition device 14. Such data includes, for example, the identification information of the vehicle 2, the parameters of the camera 11 such as the installation height of the camera 11, the shooting direction and the angle of view, and a parameter set for identifying a classifier for detecting features from an image. The memory 22 may also store the image received from the camera 11 and the positioning information received from the GPS receiver 12 for a certain period of time. The memory 22 further stores information indicating an area (hereinafter, sometimes referred to as a collection target area) that is designated in the instruction to collect feature data and is to be the target of the generation and collection of feature data. The memory 22 may also store computer programs for implementing each process executed by the processor 23.

[0026] The processor 23 has one or more CPUs (Central Processing Units) and their peripheral circuits. The processor 23 may further have other arithmetic circuits such as a logic arithmetic unit, a numerical arithmetic unit, or a graphic processing unit. The processor 23 stores images received from the camera 11 and positioning information received from the GPS receiver 12 in the memory 22. Furthermore, the processor 23 executes processes related to the generation of feature data at predetermined intervals (for example, every 0.1 to 10 seconds) while the vehicle 2 is traveling.

[0027] As a process related to the generation of feature data, the processor 23 determines whether or not the vehicle position of the vehicle 2 represented by the positioning information received from the GPS receiver 12 is included in the collection target area. If the vehicle position is included in the collection target area, the processor 23 generates feature data based on the image received from the camera 11.

[0028] For example, the processor 23 detects features shown in the input image (hereinafter, sometimes simply referred to as an input image) by inputting the image received from the camera 11 to a classifier that has been trained in advance to detect features to be detected. The processor 23 then generates information representing the type of the detected features as feature data. The processor 23 may use, as such a classifier, a deep neural network (DNN) that has been trained in advance to detect features shown in the input image from the input image. As such a DNN, for example, a DNN having a convolutional neural network (CNN) type architecture such as Single Shot MultiBox Detector (SSD) or Faster R-CNN is used. Alternatively, the processor 23 may use, as such a classifier, a DNN having a self attention network type architecture. In this case, the classifier calculates a confidence level that represents the likelihood that a feature to be detected is shown in each region for each type of feature (for example, a lane marking, a pedestrian crossing, a stop line, etc.) in various regions on the input image. The classifier determines that a certain type of feature is represented in an area where the certainty of the feature is equal to or greater than a predetermined detection threshold. The classifier then outputs information representing an area in the input image that contains the feature to be detected (e.g., a circumscribing rectangle of the feature to be detected, hereinafter referred to as an object area) and information representing the type of feature represented in the object area. Therefore, the processor 23 may generate feature data so as to include information representing the type of feature represented in the detected object area.

[0029] In addition, the processor 23 may determine that a feature is represented in a certain object region when there is no feature of a type whose certainty is equal to or greater than the detection threshold, but the sum of the certainty values ​​of a predetermined number of types of features in descending order of the certainty is equal to or greater than the detection threshold. The processor 23 may generate feature data for the object region so as to include information representing each of the predetermined number of types in descending order of the certainty and the corresponding certainty. For example, it is assumed that the detection threshold is 0.8, and that the certainty value of a no parking sign, the certainty value of a no parking sign, the certainty value of a no parking sign, and the certainty value of a no entry sign are calculated to be 0.6, 0.3, and 0.1, respectively, for features represented in a certain object region. It is also assumed that the predetermined number is 2. In this case, the sum (0.9) of the certainty value of the no parking sign and the certainty value of the no parking sign is equal to or greater than the detection threshold. Therefore, the processor 23 includes information representing a no parking sign and a no parking sign in the feature data as information representing the type of feature represented in the object region. Furthermore, the processor 23 may include the confidence factor of 0.6 calculated for the no parking sign and the confidence factor of 0.3 calculated for the no parking or stopping sign in the feature data.

[0030] Furthermore, the processor 23 identifies the position of the feature in real space represented in the feature data, and includes information representing the position in the feature data. Here, the position of each pixel on the image corresponds one-to-one to the direction from the camera 11 to the object represented in that pixel. Therefore, the processor 23 estimates the position of the feature based on the direction from the camera 11 corresponding to the center of gravity of the object area in which the feature is represented, the position of the vehicle 2 at the time of generating the image in which the feature is detected, the traveling direction, and the parameters of the camera 11 (shooting direction, angle of view, installation height, etc.). At that time, the processor 23 can set the position represented by the positioning information received from the GPS receiver 12 at the timing closest to the time of generating the image used to generate the feature data as the vehicle position of the vehicle 2. Alternatively, when an ECU (not shown) estimates the vehicle position of the vehicle 2, the processor 23 may obtain information representing the estimated vehicle position of the vehicle 2 from the ECU via the communication interface 21. Furthermore, the processor 23 may obtain information representing the traveling direction of the vehicle 2 from the ECU (not shown). Alternatively, the processor 23 may estimate the position of the feature represented in the feature data by so-called Structure from Motion (SfM). In this case, the processor 23 associates object regions in which the same feature is represented between two images obtained at different times using optical flow. The processor 23 can then estimate the position of the feature by triangulation based on the position and traveling direction of the vehicle 2 when each of the two images was obtained, the parameters of the camera 11, and the position of the object region in each image.

[0031] The processor 23 includes the latitude and longitude representing the position of the feature represented in the feature data in the feature data as information representing the position of the feature represented in the feature data.

[0032] The processor 23 may further include identification information of the vehicle 2 in the feature data. The processor 23 may also include information used to estimate the position of the feature, such as parameters of the camera 11 and the position of the feature on the image, in the feature data. The processor 23 may also include the position of the vehicle 2 at the time of generating the feature data, the traveling direction, and the reception strength of the GPS signal used to measure the vehicle position of the vehicle 2, which were used when estimating the position of the feature, in the feature data. Then, every time the processor 23 generates feature data, it outputs the generated feature data to the wireless communication terminal 13 via the communication interface 21. As a result, the feature data is transmitted to the server 3. The processor 23 may transmit information used to estimate the position of the feature together with the identification information of the vehicle 2 to the server 3 via the wireless communication terminal 13, separately from the feature data.

[0033] According to a modified example, the processor 23 may treat the image itself generated by the camera 11 (hereinafter referred to as the entire image) or a partial image obtained by cutting out an area showing the road surface from the entire image as the feature data. In this case, the processor 23 may also include in the feature data the position and traveling direction of the vehicle 2 and parameters of the camera 11 at the time of generating the feature data, so that the server 3 can detect features and their positions from the entire image or the partial image.

[0034] Furthermore, when the processor 23 has received a collection stop instruction from the server 3 for the collection target area, the processor 23 does not generate feature data for features within the collection target area.

[0035] Next, the server 3, which is an example of a map updating device, will be described. 4 is a hardware configuration diagram of a server 3 which is an example of a map updating device. The server 3 has a communication interface 31, a storage device 32, a memory 33, and a processor 34. The communication interface 31, the storage device 32, and the memory 33 are connected to the processor 34 via signal lines. The server 3 may further have input devices such as a keyboard and a mouse, and a display device such as a liquid crystal display.

[0036] The communication interface 31 is an example of a communication unit, and has an interface circuit for connecting the server 3 to the communication network 4. The communication interface 31 is configured to be able to communicate with the vehicle 2 via the communication network 4 and the wireless base station 5. That is, the communication interface 31 passes the feature data received from the vehicle 2 via the wireless base station 5 and the communication network 4 to the processor 34. The communication interface 31 also transmits the collection instruction and collection stop instruction received from the processor 34 to the vehicle 2 via the communication network 4 and the wireless base station 5.

[0037] The storage device 32 is an example of a storage unit, and includes, for example, a hard disk device or an optical recording medium and an access device therefor. The storage device 32 stores various data and information used in the map update process. For example, the storage device 32 stores map information to be generated or updated, the positions and types of individual registered features shown in the map information, and identification information of each vehicle 2. The storage device 32 also stores feature group information. Furthermore, the storage device 32 stores feature data received from each vehicle 2. Furthermore, the storage device 32 may store a computer program to be executed on the processor 34 for performing the map update process.

[0038] The memory 33 is another example of a storage unit and includes, for example, a non-volatile semiconductor memory and a volatile semiconductor memory. The memory 33 temporarily stores various data generated during the execution of the map update process.

[0039] The processor 34 is an example of a control unit and includes one or more central processing units (CPUs) and their peripheral circuits. The processor 34 may further include other arithmetic circuits such as a logic arithmetic unit or a numerical arithmetic unit. The processor 34 executes a map update process.

[0040] 5 is a functional block diagram of the processor 34 related to the map update process. The processor 34 has a collection instruction unit 41, an association unit 42, an update unit 43, and a collection stop instruction unit 44. Each of these units in the processor 34 is, for example, a functional module realized by a computer program that runs on the processor 34. Alternatively, each of these units in the processor 34 may be a dedicated arithmetic circuit provided in the processor 34.

[0041] The collection instruction unit 41 generates a collection instruction to instruct the vehicle 2 to collect feature data on features existing in the collection target area. Each collection target area can be any of a plurality of areas set by dividing the entire area shown on the map to be generated or updated into a grid shape at predetermined lengths (for example, several tens to several hundreds of meters). However, without being limited to this example, each collection target area may be set so that the area of ​​the collection target area becomes smaller in an area where the density of roads is higher. Alternatively, each collection target area may be set so that the area of ​​the area becomes smaller in an area where the density of a specific road structure (for example, an intersection, a junction, or a branch) is higher. Alternatively, each collection target area may be set so that each collection target area includes one road section or one intersection. For example, when map information is newly generated, the collection instruction unit 41 sets each of a plurality of areas shown on the map information as the collection target area. Alternatively, the collection instruction unit 41 may set an area where a predetermined period has passed since the previous update, among a plurality of areas shown on the map information to be updated, as the collection target area. Alternatively, the collection instruction unit 41 may set the collection target area according to information specifying the collection target area input via an input device. Furthermore, when the collection instruction unit 41 receives construction information indicating a point where construction has been performed from a traffic information server, the collection instruction unit 41 may set the area including the point where the construction has been performed as the collection target area.

[0042] The collection instruction unit 41 generates a collection instruction including information for identifying a collection target area. The collection instruction unit 41 then transmits the generated collection instruction to the vehicle 2 via the communication interface 31.

[0043] Every time the association unit 42 receives feature data from the vehicle 2 via the wireless base station 5, the communication network 4, and the communication interface 31, the association unit 42 stores the received feature data in the memory 33 or the storage device 32. In addition, the association unit 42 associates the feature represented in the received feature data with any of the individual registered features represented in the map information.

[0044] To this end, the association unit 42 identifies registered features that are located within a predetermined distance (e.g., several meters to tens of meters) from the location of the feature represented in the received feature data, among the individual registered features represented in the map information.The association unit 42 then associates the identified registered features with registered features of the same type as the feature represented in the received feature data, or with registered features of a type that belongs to the same group.In this case, the association unit 42 refers to feature group information to identify registered features that belong to the same group as the type of feature represented in the feature data, i.e., registered features of a similar type.

[0045] FIG. 6 is a diagram showing an example of feature group information. As shown in a table 600, which is an example of feature group information, each type of feature is grouped into three or more hierarchical levels. The lower the hierarchical level of a group, the higher the similarity between the types belonging to that group. For example, regarding lane markings of road markings, both the white solid line and the yellow solid line, which are examples of features, are included in the group "solid line". Also, both the white dashed line and the yellow dashed line, which are other examples of features, are included in the group "dashed line". Furthermore, both the group "solid line" including the white solid line and the yellow solid line and the group "dashed line" including the white dashed line and the yellow dashed line are included in the group "demarcation line". Furthermore, both the boundary line by a curb / wall and the boundary line by a structure, which are other examples of features, are included in the group "road boundary line". And both the group "demarcation line" and the group "road boundary line" are included in the group "continuous feature". The feature type groups may be hierarchically organized into two levels, or into four or more levels.

[0046] 7(a) to 7(c) are diagrams showing an example of the association between the features shown in the collected feature data and the registered features. In the example shown in FIG. 7(a), the type of feature 701 shown in the feature data is a no parking road sign, and a no parking road sign is shown as the registered feature 702 in the map information within a predetermined distance D from the position of the feature 701. As shown in the table 600 in FIG. 6, the no parking road sign and the no parking road sign belong to the same group "road sign". Therefore, the association unit 42 associates the no parking road sign, which is the feature 701 shown in the feature data, with the no parking road sign, which is the registered feature 702.

[0047] In the example shown in Fig. 7(b), the type of feature 711 shown in the feature data is a stop line, and a white solid line is shown in the map information as a registered feature 712 within a predetermined distance D from the position of the feature 711. However, the map information does not show any registered features other than the registered feature 712 within the predetermined distance D from the position of the feature 711. Furthermore, with reference to table 600, the stop line and the white solid line belong to different groups. Therefore, the association unit 42 determines that there is no registered feature that corresponds to the feature 711 shown in the feature data, and does not associate the feature 711 with any registered feature.

[0048] Note that there may be a plurality of registered features belonging to the same group as the type of the feature within a predetermined distance from the position of the feature represented in the feature data. In such a case, the association unit 42 associates the feature represented in the feature data with the registered feature that is closest to the position of the feature represented in the feature data among the plurality of registered features. Alternatively, the association unit 42 may refer to the feature group information and search for registered features of the same type as the type of the feature represented in the feature data in order of increasing similarity, that is, from the lower group, among the plurality of registered features, and associate them. Furthermore, if the plurality of registered features include a registered feature of the same type as the feature represented in the feature data, the association unit 42 may preferentially associate the feature represented in the feature data with the registered feature of the same type. By performing such association, the association unit 42 can improve the possibility that the feature represented in the feature group can be appropriately associated with the corresponding registered feature represented in the map information.

[0049] In the example shown in FIG. 7(c), the type of feature 721 represented in the feature data is a road sign for no parking. Furthermore, within a predetermined distance D from the position of the feature 721, the map information shows a registered feature 722 which is a road sign for no parking and a registered feature 723 which is a traffic light type 1. Referring to the table 600, both the road sign for no parking and the road sign for no parking belong to the group "road sign" which is the lowest group. In contrast, the road sign for no parking and the traffic light type 1 belong to the group "three-dimensional feature" which is one layer above the lowest group. Therefore, the association unit 42 associates the road sign for no parking, which is the feature 721 represented in the feature data, with the registered feature 722 which is a road sign for no parking, instead of the registered feature 723 which is a traffic light type 1.

[0050] Alternatively, when there are multiple registered features that belong to the same group as the type of the feature within a predetermined distance from the position of the feature represented in the feature data, the association unit 42 may associate the feature represented in the feature data with each of the registered features. In this case, the association unit 42 may set a reliability indicating the accuracy of the association for each registered feature that is associated with the feature represented in the feature data. For example, the association unit 42 may set a higher reliability for a registered feature that is in the same group as the feature represented in the feature data and is in a lower hierarchical level. For example, in FIG. 7(c), the association unit 42 sets the reliability of association to 0.7 for the registered feature 722, which is a road sign prohibiting parking and stopping and is in the same group as the feature 721, which is a road sign prohibiting parking, and is in a lower hierarchical level. In contrast, the association unit 42 may set the reliability of association to 0.3 for the registered feature 723, which is a signal type 1 that is in the same group as the feature 721, which is a road sign prohibiting parking.

[0051] Furthermore, it is preferable that the matching unit 42 sets a higher reliability when matching a feature represented in the feature data with a registered feature of the same type as the feature's type than a higher reliability when matching the feature with a registered feature of a different type but belonging to the same group.

[0052] Furthermore, as described above, the feature data may include information representing two or more feature types and the confidence level for each type. In such a case, the association unit 42 may search for registered features of a type that belongs to the same group as the type in order of the highest confidence level, and associate the registered features with the features represented in the feature data. That is, the association unit 42 executes the above-mentioned association process in order of the highest confidence level. When a feature of any type is associated with a registered feature, the association unit 42 ends the association process for the feature data. Then, the association unit 42 sets the confidence level for the type of feature associated with the registered feature as the reliability of the association with the registered feature. In this case, the association unit 42 may associate the feature of the type with the registered feature only when the confidence level for the type of interest is equal to or higher than a predetermined threshold value. For example, assume that one piece of feature data includes information indicating two types of features, a no parking sign and a no parking / stopping sign, and a certainty factor of 0.6 for the no parking sign and a certainty factor of 0.3 for the no parking / stopping sign. Assume that a registered feature, which is a road sign, is registered in the map information within a predetermined distance from the position of the feature represented in the feature data. In this case, the no parking sign with the higher certainty factor and the registered feature belong to the same group, so the registered feature is associated with the no parking sign, and the reliability of the association is set to 0.6. By performing such association, the association unit 42 can improve the possibility of appropriately associating the features represented in the feature group with the corresponding registered feature represented in the map information.

[0053] Alternatively, the association unit 42 may determine, among the two or more types of features represented in the feature data, a type that belongs to the same group at the lowest level as the type of a registered feature within a predetermined distance from the position of the feature, as the type of feature to be associated with the registered feature. In this case, the association unit 42 may determine, as the reliability of the association, the confidence level corresponding to the type of feature associated with the registered feature.

[0054] The association unit 42 stores in the memory 33, for each piece of received feature data, the results of association of features represented in the feature data.

[0055] The update unit 43 updates at least one of the position and the type of each registered feature that is represented in the map information and is within the collection target area, based on the features represented in each of a predetermined number (for example, 1) or more pieces of feature data associated with the registered feature. Hereinafter, a feature represented in feature data associated with any registered feature may be simply referred to as a feature.

[0056] For example, the update unit 43 refers to each feature associated with the registered feature of interest, which is stored in the memory 33. The update unit 43 then updates the position of the registered feature with an average position obtained by averaging the positions of each feature associated with the registered feature of interest. In addition, when the reliability of the association is set for each feature, the update unit 43 may update the position of the registered feature with an average position obtained by weighting the positions of each feature associated with the registered feature of interest with the reliability. That is, the update unit 43 increases the weighting coefficient of the position of a feature with a higher reliability. In this way, the update unit 43 uses the position of the feature weighted by the reliability to update the position of the corresponding registered feature, thereby making it possible to more appropriately update the position of the registered feature. Furthermore, when the feature data associated with the registered feature of interest includes the confidence level of the type of feature, the update unit 43 may update the position of the registered feature with an average position obtained by weighting the positions of each feature associated with the registered feature of interest with the confidence level.

[0057] Furthermore, the updating unit 43 may update the type of the registered feature of interest with the type that is the most frequent among the types of each feature associated with the registered feature of interest. When a reliability of association is set for each feature, the updating unit 43 may calculate the sum of the reliability for each type of feature associated with the registered feature of interest and obtain the maximum value of the sum of the reliability as the mode. Furthermore, when the feature data associated with the registered feature of interest includes a certainty of the type of feature, the updating unit 43 may calculate the sum of the certainty for each type of feature associated with the registered feature of interest and obtain the maximum value of the sum of the certainty as the mode.

[0058] Furthermore, when there is no feature associated with the registered feature of interest, the update unit 43 may determine that the registered feature of interest has been removed and delete the registered feature from the map information. Furthermore, the update unit 43 may determine that a feature represented in feature data that has not been associated with any registered feature has been newly established and add the type and position of the feature to the map information. In this case, the update unit 43 may determine that, among the features represented in a plurality of feature data that have not been associated with any registered feature, two or more features that are a predetermined distance or less apart and belong to the same group are the same feature. Then, for two or more features that have been determined to be the same feature, the update unit 43 may add only one of the features to the map information. In this case, the update unit 43 may set the average position of the two or more features as the position of the feature to be registered in the map information, and set the type that is the most frequent type among the types of the two or more features as the type of the feature to be registered in the map information.

[0059] When the number of feature data collected for the collection target area reaches the planned number of data to be collected, or when the planned collection period ends, the collection stop instruction unit 44 stops collecting feature data for the collection target area. Then, the collection stop instruction unit 44 creates a collection stop instruction including information representing the stop of collection for the collection target area, and transmits the created collection stop instruction to the vehicle 2 via the communication interface 31. This stops the collection of feature data related to features located in the collection target area.

[0060] 8 is an operation flowchart of the map update process in the server 3. When the number of feature data collected for the collection target area reaches the planned number of data to be collected or the collection period ends, the processor 34 of the server 3 may execute the map update process according to the operation flowchart shown below.

[0061] The association unit 42 of the processor 34 identifies, for each collected feature data, a registered feature located within a predetermined distance from the position of the feature represented in the feature data among the individual registered features represented in the map information (step S101). Then, for each feature data, the association unit 42 refers to feature group information and associates the feature represented in the feature data with a registered feature of the same type as the feature represented in the feature data or a type belonging to the same group among the identified registered features (step S102). Furthermore, the update unit 43 of the processor 34 updates at least one of the position or type of the registered feature represented in the map information and within the collection target area based on the position or type of the feature represented in each feature data associated with the registered feature (step S103). Then, the processor 34 ends the map update process. Note that the association unit 42 may execute the processes of steps S101 and S102 every time it receives feature data from the vehicle 2.

[0062] As described above, this map updating device makes it possible to associate features represented in the collected feature data not only with registered features of the same type represented in the map information, but also with registered features of similar types belonging to the same group. Therefore, this map updating device can correct the position of the registered feature represented in the map information using the collected feature data even if the vehicle side generating the feature data erroneously recognizes the type of the feature or the type of the registered feature represented in the map information is incorrect. Furthermore, this map updating device can correct the type of the registered feature represented incorrectly in the map information based on the type of the feature represented in the feature data. In this way, this map updating device can appropriately update the map information using the collected feature data.

[0063] According to a modified example, the position of each registered feature shown in the map information may be represented by a probability distribution that represents the probability that the registered feature exists at each position. The probability distribution may be, for example, a Gaussian distribution. In this case, the association unit 42 determines that the registered feature exists within a predetermined distance from the position of the feature represented in the feature data when the position at which the registered feature exists most likely in the probability distribution is included within a predetermined distance from the position of the feature represented in the feature data. Furthermore, the update unit 43 may update the probability distribution of the position of each registered feature shown in the map information by performing maximum likelihood estimation based on the position of the feature in each feature data associated with the registered feature. When updating the probability distribution of the position of the registered feature, the update unit 43 may use the reliability of the association or the confidence obtained at the time of feature detection as a weighting coefficient, as in the above embodiment.

[0064] Furthermore, the update unit 43 may update the probability distribution of the position of each registered feature represented in the map information by performing Bayesian estimation based on the position of the feature in each feature data associated with the registered feature. In this case, for a registered feature associated with a feature represented in the received feature data, a range in which the registered feature may exist is divided in advance into a plurality of grid-like sections. Then, for each section, a probability that the registered feature is located is set according to the probability distribution before the update. Note that the same probability may be set for each section as an initial value of the probability, or a higher probability may be set for a section in which the registered feature is more likely to exist. When the feature represented in the received feature data is associated with a registered feature of interest, the update unit 43 updates the probability of each section set for the registered feature so that the probability of the section including the position of the feature represented in the received feature data becomes higher. Alternatively, the update unit 43 may update the probability of each section set for the registered feature so that the probability of each section within a predetermined range from the position of the feature represented in the received feature data becomes higher. At this time, the update unit 43 may increase the rate of increase in the probability for a section closer to the position of the feature. Alternatively, the update unit 43 may set, for each section, a probability distribution centered on the section and having the variance-covariance matrix of the probability distribution before the update. For each section, the update unit 43 calculates a posterior probability that a feature exists in the section with respect to the position of the feature represented in the received feature data, based on the probability distribution of the section. For each section, the update unit 43 sets the calculated posterior probability as the post-update probability of the section (i.e., the prior probability of the section at the time of the next update). Then, the update unit 43 calculates an updated probability distribution for the position of the feature by approximating the probability of each section with a normal distribution. In this case, the feature data used to update the probability distribution may be discarded, so that the update unit 43 can simplify the management of the feature data and reduce the memory capacity for storing the feature data.

[0065] Alternatively, the update unit 43 may set multiple candidates of the probability distribution for the position of each registered feature. In this case, each candidate can be a normal distribution represented by the average value of the position and the variance covariance matrix. When the update unit 43 associates a feature represented in the received feature data with a registered feature of interest, it calculates the posterior probability of each candidate for the registered feature with respect to the position of the feature represented in the received feature data. The update unit 43 then sets the posterior probability as the prior probability of each candidate for the next update. The update unit 43 sets the normal distribution corresponding to the candidate with the maximum prior probability as the probability distribution for the position of the registered feature. In this case as well, the update unit 43 can simplify the management of feature data and reduce the memory capacity for storing feature data.

[0066] Also, a single vehicle 2 may collect a plurality of feature data representing a plurality of features existing in a road section that is continuous to a certain extent. In such a case, the positions of the features represented in each of the plurality of feature data may deviate from the actual positions of the features in substantially the same direction and by substantially the same distance due to the positioning accuracy of the vehicle 2 or the like. Therefore, according to another modification, the association unit 42 corrects the positions of the multiple features represented in each of the multiple feature data for the continuous road section received from a single vehicle 2. In this case, the association unit 42 corrects the positions of the features of each piece of feature data so as to minimize the sum of squares of the errors between the positions of the features of each piece of feature data and the positions of the corresponding registered features represented in the map information, and then associates the corrected positions with the registered features. In this case, the association unit 42 may refer to the feature group information and identify the corresponding registered features by assuming that the registered features that belong to the same group as the type of feature included in each piece of feature data correspond to the corresponding feature.

[0067] When the feature data received by the server 3 from the vehicle 2 includes a full image or a partial image, in the above embodiment and each of the modified examples, the processor 34 of the server 3 may execute the processing of the processor 23 of the data acquisition device 14. That is, the processor 34 may detect features from the full image or the partial image, and estimate the position of the detected features.

[0068] Furthermore, a computer program that causes a computer to realize the functions of each unit of the processor of the map updating device according to each of the above embodiments or modifications may be provided in a form stored in a computer-readable recording medium. Note that the computer-readable recording medium may be, for example, a magnetic recording medium, an optical recording medium, or a semiconductor memory.

[0069] As described above, those skilled in the art can make various modifications to the embodiments within the scope of the present invention. [Explanation of symbols]

[0070] 1. Geophysical data collection system 2 Vehicles 11 Camera 12 GPS receiver 13 Wireless communication terminals 14 Data acquisition device 21 Communication Interface 22 Memory 23 Processors 3 Server (map update device) 31 Communication Interface 32 Storage Device 33 Memory 34 processors 41 Collection Instructions Division 42 Mapping section 43 Update section 44 Collection Suspension Instructions 4. Communication Network 5. Radio base station

Claims

1. a storage unit that stores the position and type of each registered feature shown in the map information, and feature group information that represents two or more similar types of features as the same group among a plurality of types of features; a correspondence unit that identifies, among the individual registered features represented in the map information, one or more registered features that are located within a predetermined distance from a position of the feature represented in feature data received from the vehicle via a communication unit, and refers to the feature group information to correspond the feature represented in the feature data to a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature; an update unit that updates at least one of a position or a type of at least one of the registered features represented in the map information based on a position or a type of the feature represented in a predetermined number of the feature data associated with the registered feature; having the feature group information represents groups of two or more levels having different degrees of similarity between the types of the features to which the features belong; The matching unit searches for and matches registered features of a type that belongs to the same group as the type of the feature represented in the feature data, among the one or more identified registered features, in order of group with the highest degree of similarity.

2. The map updating device described in claim 1, wherein the updating unit sets a weighting coefficient for at least any of the individual registered features represented in the map information, the higher the degree of similarity of the group that is the same as the type of the registered feature, for the feature represented in each of the predetermined number of feature data associated with the registered feature, and updates the position of the registered feature to a position obtained by weighting the positions of the features represented in each of the predetermined number of feature data using the corresponding weighting coefficient.

3. A storage unit that stores the positions and types of each registered feature represented in map information, and feature group information that represents two or more types of features that are similar to each other as the same group among multiple types of features; a correspondence unit that identifies, among the individual registered features represented in the map information, one or more registered features that are located within a predetermined distance from a position of the feature represented in feature data received from the vehicle via a communication unit, and refers to the feature group information to correspond the feature represented in the feature data to a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature; an update unit that updates at least one of a position or a type of at least one of the registered features represented in the map information based on a position or a type of the feature represented in a predetermined number of the feature data associated with the registered feature; having The feature data further includes a degree of certainty for each of two or more types of the features represented in the feature data, the degree of certainty indicating a likelihood that the features are of the respective types; The matching unit searches for and matches registered features of a type that belongs to the same group as the type of the feature represented in the feature data, in descending order of the degree of certainty for the feature.

4. A map update device, among the individual registered features represented in the map information, identifies one or more registered features located within a predetermined distance from a position of a feature represented in the feature data received from the vehicle via a communication unit; the map updating device refers to feature group information that represents two or more mutually similar types of the plurality of types of the feature as the same group, and associates the feature represented in the feature data with a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature, among the one or more registered features identified; the map updating device updates at least one of a position or a type of at least any of the individual registered features represented in the map information based on a position or a type of the feature represented in a predetermined number of the feature data associated with the registered feature; Including, the feature group information represents groups of two or more levels having different degrees of similarity between the types of the features to which the features belong; The step of associating the one or more registered features includes searching for registered features of a type that belongs to the same group as the type of the feature represented in the feature data, in order from the group with the highest similarity, among the one or more identified registered features, and associating the registered features with the one or more registered features. How to update maps.

5. A map update device identifies, among the individual registered features represented in the map information, one or more registered features that are located within a predetermined distance from a position of a feature represented in the feature data received from the vehicle via a communication unit, the map updating device refers to feature group information that represents two or more mutually similar types of the plurality of types of the feature as the same group, and associates the feature represented in the feature data with a registered feature of a type that is the same as the type of the feature represented in the feature data or that belongs to the same group as the type of the feature, among the one or more registered features identified; the map updating device updates at least one of a position or a type of at least any of the individual registered features represented in the map information based on a position or a type of the feature represented in a predetermined number of the feature data associated with the registered feature; Including, The feature data further includes a degree of certainty for each of two or more types of the features represented in the feature data, the degree of certainty indicating a likelihood that the features are of the respective types; The step of associating the feature with the registered feature includes searching for and associating a registered feature of a type that belongs to the same group as the feature represented in the feature data in descending order of the degree of certainty of the feature. How to update maps.

6. Among the individual registered features represented in the map information, one or more registered features located within a predetermined distance from a position of the feature represented in the feature data received from the vehicle via the communication unit are identified; referring to feature group information that represents two or more similar features as the same group among the plurality of features' types, the feature represented in the feature data is associated with a registered feature of a type that is the same as the feature's type represented in the feature data or that belongs to the same group as the feature's type among the one or more identified registered features; updating at least one of the position or the type of at least any of the individual registered features represented in the map information based on the position or the type of the feature represented in a predetermined number of the feature data associated with the registered feature; Have the computer do this, the feature group information represents groups of two or more levels having different degrees of similarity between the types of the features to which the features belong; The step of associating the one or more registered features includes searching for registered features of a type that belongs to the same group as the type of the feature represented in the feature data, in order from the group with the highest similarity, among the one or more identified registered features, and associating the registered features with the one or more registered features. A computer program for updating maps.

7. Among the individual registered features represented in the map information, one or more registered features located within a predetermined distance from a position of a feature represented in the feature data received from the vehicle via a communication unit are identified; referring to feature group information that represents two or more similar features as the same group among the plurality of features' types, the feature represented in the feature data is associated with a registered feature of a type that is the same as the feature's type represented in the feature data or that belongs to the same group as the feature's type among the one or more identified registered features; updating at least one of the position or the type of at least any of the individual registered features represented in the map information based on the position or the type of the feature represented in a predetermined number of the feature data associated with the registered feature; Have the computer do this, The feature data further includes a degree of certainty for each of two or more types of the features represented in the feature data, the degree of certainty indicating a likelihood that the features are of the respective types; The step of associating the feature with the registered feature includes searching for and associating a registered feature of a type that belongs to the same group as the feature represented in the feature data in descending order of the degree of certainty of the feature. A computer program for updating maps.

Images