JP7039536B2 - Map information processing system - Google Patents

Description

The present invention relates to a map information processing system that determines whether or not the road shape of map information has changed from the trajectory information of a plurality of moving objects.

Currently, road information is managed in units of administrative divisions such as countries, prefectures, cities, and wards, and information on changes in road shape (hereinafter also referred to as "road shape changes") is not centralized. In order to grasp the change in road shape in order to create a road map, it is necessary to investigate where and how the new road was formed for each individual road.

Therefore, based on the digital map issued by the Geographical Survey Institute, for example, and the probe data (Probe Data: trajectory information) acquired from the vehicle terminal, it is determined how the road shape has changed, and the road is determined. A method of updating the map is conceivable.

As a device for generating this kind of map information, there is known a configuration in which a road connection condition is correctly estimated based on the position information acquired by an in-vehicle terminal and an updated map is generated (for example, Patent Document 1).

Japanese Unexamined Patent Publication No. 2017-97088

However, since the road shape is complicated and has various patterns such as when the crossroad shape is changed to the rotary shape, it may be difficult to simply generate an updated map from the map information and the position information.

Therefore, the configuration described in Patent Document 1 is not sufficient, and further improvement is required.
The present invention has been made in view of such a problem, and an object of the present invention is to make it easier to determine a change in road shape.

According to one aspect of the present disclosure , the map information processing apparatus of the present disclosure includes a locus information acquisition unit, a map information storage unit, a map information conversion unit, a difference extraction unit, and a road shape change determination unit. .. The locus information acquisition unit acquires the locus information of each of the plurality of moving bodies. The map information storage unit stores map information represented by vector data. The map information represented by the vector data includes information indicating the position and shape of the road. The map information conversion unit converts the map information from vector data to scalar data. The difference extraction unit extracts the difference data. The difference data represents the difference between the movement locus image drawn from the locus information and the map image of the road in the map information converted into the scalar data. The road shape change determination unit determines whether or not the difference data is a road shape change by machine learning.

INDUSTRIAL APPLICABILITY The present invention can more easily determine a change in road shape.

FIG. 1 is a diagram showing an example of a map information processing system of the present embodiment. FIG. 2 is a diagram showing an example of the relationship between the map information processing device of the present embodiment and the in-vehicle terminal. FIG. 3 is a functional block diagram showing an example of the functions of the map information processing apparatus of the present embodiment. FIG. 4A is a diagram showing an example of a method of drawing a moving locus image of a vehicle by a moving image drawing unit. FIG. 4B is a diagram showing an example of a method of drawing a moving locus image of a vehicle by a moving image drawing unit. FIG. 5A is a diagram showing an example of a method of extracting difference data from a movement locus image and a map image by a difference extraction unit. FIG. 5B is a diagram showing an example of a method of extracting difference data from a movement locus image and a map image by a difference extraction unit. FIG. 5C is a diagram showing an example of a method of extracting difference data from a movement locus image and a map image by a difference extraction unit. FIG. 5D is a diagram showing an example of a method of extracting difference data from a movement locus image and a map image by a difference extraction unit. FIG. 6 is a diagram showing an example of a method of determining a road shape change by machine learning. FIG. 7 is a flowchart showing an example of road shape change determination processing by the road shape change determination processing program. FIG. 8 is a diagram showing an example of road shape change determination processing. FIG. 9 is a flowchart showing an example of map information update processing by the road shape change determination processing program. FIG. 10A is a diagram showing an example of map information update processing. FIG. 10B is a diagram showing an example of map information update processing. FIG. 10C is a diagram showing an example of map information update processing. FIG. 10D is a diagram showing an example of map information update processing. FIG. 10E is a diagram showing an example of map information update processing.

The map information processing system 1 of the embodiment will be described with reference to FIG. The map information processing system 1 of the present embodiment is used to easily determine a change in road shape when creating a map. As shown in FIG. 1, the map information processing system 1 includes a map information processing device 2 and a plurality of in-vehicle terminals 3. Hereinafter, each configuration in the map information processing system 1 of the present embodiment will be described in detail.

The in-vehicle terminal 3 is mounted on a vehicle 4 traveling on a road. The in-vehicle terminal 3 is composed of, for example, a car navigation device. The map information processing device 2 and the vehicle-mounted terminal 3 are configured to be able to be connected to each other by wireless communication. In FIG. 1, a single arrow illustrates wireless communication between a vehicle 4 traveling in Japan and a map information processing device 2. In the map information processing system 1, probe data is transmitted from the vehicle-mounted terminal 3 to the map information processing device 2.

The probe data includes, for example, information on the own vehicle position (hereinafter, also referred to as “trajectory information”) and time of the vehicle 4 on which the in-vehicle terminal 3 is mounted. The probe data includes information on the travel history (attribute) of the vehicle 4, such as mileage, travel speed, acceleration, and angular velocity, in addition to information on the position and time of the own vehicle. The creation of probe data is, for example,
This is done using the in-vehicle terminal 3 mounted on the vehicle 4. The in-vehicle terminal 3 can obtain various sensor information from a GPS (Global Positioning System) module, a gyro sensor, and an acceleration sensor mounted on the vehicle 4. The information on the position of the own vehicle is calculated from, for example, GPS data output from the GPS module of the car navigation device. That is, the information on the position of the own vehicle is acquired by using the function of the satellite positioning system such as GPS.
The information on the position of the own vehicle represents the position coordinates of the vehicle 4 at a certain time.

As shown in FIG. 2, the in-vehicle terminal 3 is configured to be able to communicate with the map information processing device 2 via the communication terminal 5. The in-vehicle terminal 3 transmits probe data including trajectory information of the vehicle 4 to the map information processing device 2 via the communication terminal 5. The in-vehicle terminal 3 may be connected to the communication terminal 5 by wire or wirelessly. The in-vehicle terminal 3 may be physically configured separately from the communication terminal 5, or may be integrally configured with the communication terminal 5 built in. The communication terminal 5 is, for example, a communication unit, a mobile phone, or a smartphone. In the case of a smartphone, the communication terminal 5 is connected so as to be able to communicate with the in-vehicle terminal 3 using, for example, BLUETOOTH (registered trademark) or a Wi-Fi communication standard. In the case of a smartphone, the communication terminal 5 is connected so as to be able to communicate with the map information processing device 2 via, for example, a communication network.

The map information processing device 2 receives the probe data transmitted from each in-vehicle terminal 3 and acquires the trajectory information included in the probe data. The map information processing device 2 determines whether or not the road shape included in the map information has changed by machine learning based on the trajectory information. The details of this determination will be described later. When the map information processing device 2 determines that the road shape has changed, the map information processing device 2 updates the map information. Then, the map information processing device 2 transmits the updated map information to the in-vehicle terminal 3.

In the above-described embodiment, the in-vehicle terminal 3 is configured by a car navigation device, but the present invention is not limited to this. The in-vehicle terminal 3 is not limited to the car navigation device, and may be various terminal devices. Examples of the terminal device include a mobile phone, a smartphone, a notebook computer, or a tablet computer. That is, the in-vehicle terminal 3 may be only a smartphone brought into the vehicle 4. Further, the vehicle 4 of the present embodiment is exemplified by a four-wheeled vehicle on which the in-vehicle terminal 3 is mounted. The vehicle 4 is not limited to a four-wheeled vehicle, and may be a three-wheeled vehicle, a motorcycle, or a bicycle. In other words, the vehicle 4 is a moving body.

Next, the map information processing apparatus 2 of the present embodiment will be described with reference to FIG.
The map information processing device 2 of the present embodiment includes a map information storage unit 21, a trajectory information acquisition unit 22, a moving image drawing unit 23, a map information conversion unit 24, a difference extraction unit 25, and a road shape change determination unit. 26 and the map information updating unit 27 are included. The map information storage unit 21 stores map information. It is preferable that the map information storage unit 21 stores map information for each of a plurality of scales.

Map information is information that constitutes a map. Map information is composed of vector data. It is more preferable that the map information is composed of three-dimensional vector data. The vector data includes information indicating the position of the road and the shape of the road. More specifically, the map information has, for example, map image information indicating the shape of a road, information on nodes and links on the map image linked to the map image information, and attribute information indicating whether it is a general road or a highway. ing. Nodes indicate intersections and other nodules on the road network representation. The link indicates the road section between the nodes. In addition, the map information is composed of mesh units separated into rectangles at regular latitude / longitude intervals. Further, each mesh is composed of a plurality of layers having different scales separated by a predetermined unit. In the case of Japan, for example, the mesh can adopt the standard of the standard area mesh set by the Ministry of Internal Affairs and Communications. The standard area mesh is composed of a primary mesh, a secondary mesh, and a tertiary mesh in the order of about 1/10 of the area ratio. Further, as the mesh, a divided area mesh subdivided from the primary to tertiary mesh can be adopted as a mesh unit. When the map information is divided into mesh units, the map information has a mesh number and corresponding latitude and longitude information, respectively.

The trajectory information acquisition unit 22 includes a vehicle data reception unit 28 and a data storage unit 29. The vehicle data receiving unit 28 receives probe data transmitted from the vehicle-mounted terminals 3 of each of the plurality of vehicles 4. The data storage unit 29 stores the probe data received by the vehicle data reception unit 28. By receiving the probe data from each in-vehicle terminal 3, the locus information acquisition unit 22 can acquire the locus information of each vehicle 4 included in the probe data. The locus information is coordinates represented by latitude and longitude.

The moving image drawing unit 23 collectively draws a moving locus image based on the locus information of a plurality of vehicles 4 in a predetermined range acquired by the locus information acquisition unit 22. For the predetermined range, for example, a mesh of any scale constituting the road information is adopted.

Hereinafter, an example in which the moving image drawing unit 23 draws a moving locus image of the vehicle 4 will be described. As shown in FIG. 4A, the moving image drawing unit 23 is continuous as shown in FIG. 4A, based on the coordinates, time, traveling speed, acceleration, and angular velocity information of the trajectory information included in the plurality of probe data received by the vehicle data receiving unit 28. Acquire m group of probe data. For example, the moving image drawing unit 23 acquires a group of probe data having continuous coordinates as a group of probe data at predetermined time intervals among the probe data. The moving image drawing unit 23 acquires the coordinates of m locus information included in the group of probe data from the data storage unit 29. The moving image drawing unit 23 plots the acquired coordinates as points P1, P2, and P3 to Pm. As shown in FIG. 4B, the moving image drawing unit 23 connects the plotted points P1, P2, and P3 to Pm with a line and draws them as one moving locus image L. As a result, the moving image drawing unit 23 can generate a scalar-format moving locus image from vector-format probe data including time, mileage, traveling speed, acceleration, and angular velocity information in addition to the locus information. ..

Further, when the vehicle data receiving unit 28 receives the probe data, the moving image drawing unit 23 repeatedly draws the moving locus image. When the moving image drawing unit 23 superimposes and draws a plurality of moving locus images, it calculates the median value or the mean square value of each moving locus image and draws the superposed moving locus images in the width direction. A line connecting the center values (center coordinates) can be drawn as an averaged movement locus image. As a result, the moving image drawing unit 23 suppresses the fluctuation of the coordinates of the trajectory information due to the error of GPS data due to multipath and the communication failure. The moving image drawing unit 23 can improve the drawing accuracy by suppressing the fluctuation of the coordinates.

The moving image drawing unit 23 may draw a moving locus image by referring to the information of the traveling speed included in the probe data and omitting the probe data whose traveling speed is equal to or higher than a predetermined speed. When the vehicle-mounted terminal 3 calculates the traveling speed from GPS data, the vehicle speed may be calculated to be 300 km or more, which is normally impossible, for example, due to an error in GPS data due to multipath or a communication failure. The moving image drawing unit 23 can suppress erroneous detection of traveling information due to an error in GPS data due to multipath or a communication failure. The moving image drawing unit 23 improves drawing accuracy by suppressing erroneous detection of traveling information.

The moving image drawing unit 23 may draw a moving locus image by referring to the information of the traveling speed included in the probe data and omitting the probe data in which the traveling speed is not continuous. As a result, the moving image drawing unit 23 can omit the probe data obtained from the in-vehicle terminal 3 of the vehicle 4 parked on the way in the parking lot, for example. The moving image drawing unit 23 can improve the drawing accuracy of the moving locus image by omitting unnecessary probe data.

The moving image drawing unit 23 can refer to the trajectory information and the traveling speed information included in the probe data, and can estimate that the probe data is moving on the expressway and that the moving image drawing unit 23 is moving on the general road. It is also possible to draw a movement trajectory image by distinguishing the probe data from each other. As a result, the map information processing apparatus 2 can set only the arbitrarily selected road type as the road shape change determination processing target. The map information processing device 2 can improve the processing speed.

Further, the moving image drawing unit 23 may draw a moving locus image based on the locus information of the vehicle 4 acquired in a predetermined unit period. The moving image drawing unit 23, for example, normalizes the data based on the locus information accumulated during a predetermined unit period. The moving image drawing unit 23 can create probe data for each unit period from the normalized data for each mesh unit. The predetermined unit period can be set, for example, for the past 30 days based on the current time point. In the present embodiment, the moving image drawing unit 23 creates the probe data of the secondary mesh as the probe data for each unit period. As described above, the map information processing apparatus 2 creates probe data based on the locus information accumulated over a predetermined period, thereby improving the drawing accuracy of the moving locus image.

The moving image drawing unit 23 may accumulate past probe data for reference. The moving image drawing unit 23 accumulates the past probe data as a reference for a specific period in advance before the probe data of the secondary mesh created based on the trajectory information acquired in a predetermined unit period is created. It can be created based on the track information. It is preferable that the past probe data is created for each mesh unit. The moving image drawing unit 23 may be set in advance as a specific period, for example, 30 days as past probe data. In the present embodiment, the moving image drawing unit 23 creates the probe data of the secondary mesh accumulated for 30 days as the past probe data.

As the first step, the moving image drawing unit 23 compares the probe data of the newly created secondary mesh with the probe data of the secondary mesh created in the past, and determines whether or not there is a difference. Can be done. Then, the moving image drawing unit 23 compares the new probe data with the map information as the second step only when there is a difference. As a result, the map information processing apparatus 2 can detect the change in the road shape in two steps. Further, the map information processing apparatus 2 targets only the probe data having a difference from the past probe data as the processing target of the difference data extraction process by the difference extraction unit 25. Since the map information processing apparatus 2 targets only the probe data having a difference from the past probe data as the processing target of the difference data, the processing target in the difference extraction unit 25 can be reduced. Since the map information processing apparatus 2 can reduce the processing targets in the difference extraction unit 25, it is possible to reduce the processing load and improve the processing speed.

In the above embodiment, the predetermined unit period is set to 1 day, 3 days, 7 days, and 30 days, but the present invention is not limited to this. The predetermined unit period may be not only a daily unit but also an hour unit, a monthly unit, or a year unit, and can be set to an arbitrary period. Further, in the above-described embodiment, the specific period is set to 30 days, but the present invention is not limited to this. The specific period can be set to any period. Further, in the above-described embodiment, the moving image drawing unit 23 is not limited to the configuration for creating probe data of the secondary mesh. The moving image drawing unit 23 may create probe data according to the primary mesh, the tertiary mesh, or another unit mesh.

The map information conversion unit 24 converts the map information in the vector format into the map information in the scalar format. Specifically, among the map information of the vector data stored in the map information storage unit 21, the map information conversion unit 24 has node information, link information, general road or highway associated with the map image information. The attribute information indicating is omitted. In other words, the map information conversion unit 24 extracts only the map image information indicating the shape of the road. The map information conversion unit 24 converts the map information from the three-dimensional vector data to the two-dimensional scalar data by using only the map image information. That is, the map information of the three-dimensional vector data is only the map image information by removing the node information, the link information, and the attribute information included in the map information. Since the map information of the three-dimensional vector data is only the map image information, it is converted into the map information of the two-dimensional scalar data having a simple data structure. As a result, the map information processing apparatus 2 can convert the movement trajectory image obtained from the probe data and the map information into data of the same simple dimension, and facilitates the extraction of the difference by the difference extraction unit 25 described later. Can be. Further, the map information conversion unit 24 converts the two-dimensional map information converted into scalar data into scalar data divided into mesh units of an arbitrary scale.

In the above-described embodiment, the map information conversion unit 24 is not limited to the case where the two-dimensional map information converted into scalar data is converted into scalar data divided into mesh units of an arbitrary scale. For example, when the road shape change determination unit 26 determines from the difference data that the probability that there is a road shape change is less than or equal to a predetermined accuracy, the map information conversion unit 24 is stored in the map information storage unit 21. Of the existing map information, map information with a large scale is converted from vector data to scalar data.

As a result, the map information processing apparatus 2 can improve the speed of determining whether or not there is a change in the road shape by roughly determining the whole at the initial stage. In other words, when the map information processing apparatus 2 determines from the difference data that the probability that there is a road shape change is a predetermined accuracy or less than a predetermined accuracy, the map information processing apparatus 2 sets the processing target to map information having a larger scale. The processing speed can be improved by limiting the processing target. As a result, the map information processing apparatus 2 can further improve the determination accuracy.

The difference extraction unit 25 can extract the difference data between the movement locus image drawn by the moving image drawing unit 23 and the map image of the map information converted into scalar data by the map information conversion unit 24. It is configured.

In the map information processing device 2 of the present embodiment, the map information storage unit 21 and the data storage unit 29 can be configured by, for example, a hard disk drive or a memory of a semiconductor memory. A program for driving a CPU (Central Processing Unit) may be stored in the memory. The CPU can function the moving image drawing unit 23, the map information conversion unit 24, the difference extraction unit 25, the road shape change determination unit 26, and the map information update unit 27 by executing the program stored in the memory. It is configured to be able to. The vehicle data receiving unit 28 can be configured by an appropriate communication module.

Hereinafter, an example of extracting the difference data by the difference extraction unit 25 will be described with reference to FIGS. 5A to 5D.
The difference extraction unit 25 acquires a movement locus image drawn by the movement image drawing unit 23 and a map image of map information converted into scalar data by the map information conversion unit 24. FIG. 5A illustrates a movement locus image drawn by the moving image drawing unit 23. FIG. 5B illustrates a map image of map information converted into scalar data. Next, as shown in FIG. 5C, the difference extraction unit 25 creates an image of the composite data obtained by synthesizing the acquired movement locus image and the map image. As a result of creating an image of the composite data, the difference extraction unit 25 uses an image that does not overlap between the movement locus image shown in FIG. 5A and the map image shown in FIG. 5B as an image of the difference data, as shown in FIG. 5D. Extract. As a result, it is possible to eliminate the fluctuation of the movement locus from the movement locus images and extract only the image of the movement locus that protrudes from the map image. FIG. 5D illustrates an image of the difference data extracted by the difference extraction unit 25. The image of the difference data is considered to represent the difference between the road at the time when the map information is generated and the current road. That is, the image of the difference data is considered to represent the change in road shape. In the examples shown in FIGS. 5A to 5D, a new road may have been formed in the circled area in FIG. Next, the map information processing apparatus 2 determines whether or not the extracted difference data is a road shape change, including the circled portion in FIG.

The road shape change determination unit 26 determines by machine learning whether or not the difference data extracted by the difference extraction unit 25 is a road shape change. Deep learning is used as machine learning. Specifically, the road shape change determination unit 26 performs machine learning using the difference data extracted in the past as teacher data. The road shape change determination unit 26 determines whether or not the newly extracted difference data is a road shape change based on the result of machine learning.

Next, with reference to FIG. 6, an example of a method in which the road shape change determination unit 26 determines the road shape change by deep learning will be described. Deep learning is a kind of machine learning method by a multi-layer neural network 60. The neural network 60 has input data and output data, and internal arithmetic processing is performed based on a plurality of artificial neurons. The neural network 60 used in machine learning includes three layers. The three layers are illustrated as an input layer 61, an intermediate layer 62, and an output layer 63. The intermediate layer 62 is also called a hidden layer and may include two or more layers. The intermediate layer 62 becomes unlearned if the number of layers is too small. If the number of layers is too large, the intermediate layer 62 becomes overfitted. The road shape change determination unit 26 may appropriately set the number of layers in order to determine whether or not the difference data is a road shape change. The artificial neuron outputs the sum of the output of the previous layer multiplied by the parameter. The output data of the artificial neuron is controlled by the activation function, and non-linearity is added. As the activation function for machine learning used in this embodiment, for example, a softmax function, a sigmoid function, or a Gaussian function can be adopted.

Since the neural network 60 performs machine learning, teacher data is first given to the input layer 61 as input data. The teacher data is, for example, difference data extracted in the past by the difference extraction unit 25. The road shape change determination unit 26 processes the teacher data by the input layer 61, the intermediate layer 62, and the output layer 63. That is, the road shape change determination unit 26 dynamically generates and learns the optimum feature amount for the input difference data, and performs forward propagation that is arithmetically processed by information propagation in the forward direction. In FIG. 6, the direction of forward propagation is indicated by a thick single arrow. The output result represents a prediction result of whether the input difference data is an image of a road shape change or noise.

Further, in the case of learning, the road shape change determination unit 26 gives information on whether the output result is an image of the road shape change or noise, and performs arithmetic processing by information propagation in the reverse direction. Propagate. In FIG. 6, the direction of back propagation is indicated by a thick single arrow. In machine learning, the road shape change determination unit 26 evaluates an output error between the input data used for learning as teacher data and the output data. It is preferable that the road shape change determination unit 26 sequentially optimizes the parameters of each layer and each node in machine learning from the output error by back propagation.

By this learning, the road shape change determination unit 26 can gradually bring the parameters of each layer closer to the optimum values. Then, when the road shape change determination unit 26 inputs the difference data extracted by the difference extraction unit 25 into the input layer 61, the difference data is changed to the road shape by using the parameters adjusted based on the result of machine learning. It is possible to determine whether or not the image is an image.

For example, as shown in FIG. 6, when the road shape change determination unit 26 sequentially propagates n difference data d1 and d2 to dn as teacher data, information on M output data y1 to yM is transmitted to the output layer 63. Is obtained. In the present embodiment, the output data represents a predicted value for predicting whether the input difference data is an image of a road shape change or another image.

The road shape change determination unit 26 is given information on M correct answer data t1 to tM indicating whether it is an image of the road shape change or noise with respect to the obtained output data y1 to yM. When the information of tM is given from the correct answer data t1 and the back propagation is performed, the road shape change determination unit 26 performs machine learning to adjust the sequential parameter to the optimum value. In other words, the road shape change determination unit 26 evaluates the deviation between the output data and the correct answer data by back propagation, and optimizes the parameters. The software used for machine learning is, for example, OpenCV, Numpy.
, Matplotlib, or Chainer can be adopted.

The map information updating unit 27 updates the map information stored in the map information storage unit 21. Specifically, the map information update unit 27 converts the map information of the vector data from the map information converted into the scalar data corresponding to the difference data determined to be the road shape change by the road shape change determination unit 26. Generate. Then, the map information updating unit 27 updates the map information stored in the map information storage unit 21 with the generated map information. The process of updating the map information will be described in the map information update process described later.

By the way, in the map information processing device, it is conceivable to simply discriminate the change in the road shape from the probe data and the digital map by machine learning.
However, in general, digital map data is a three-dimensional data having map image information indicating the shape of a road, information on nodes and links on the map image linked to the map image information, attribute information indicating whether it is a general road or a highway, and the like. It is composed of vector data.

Therefore, when the map information processing device detects a change in the road shape newly created by using machine learning, the map information processing device simply processes the three-dimensional vector data of the digital map as the processing target of the machine learning. The amount of information to be processed may be too large and the burden of information processing may become too large. Further, since the map information processing device has a large amount of information to be compared as a machine learning object, it may be difficult to determine which information is the information to be compared.

Since the map information processing device 2 of the present embodiment creates a road map by machine learning, it is easy to determine the road shape change by simplifying the processing target as much as possible.
Moreover, since the shape of the earth is an oblate spheroid rather than a true sphere, the vertical and horizontal lengths of the mesh may differ depending on the location. When performing machine learning, the map information processing apparatus 2 of the present embodiment preferably performs normalization in which the vertical and horizontal lengths of each mesh are unified in advance. The map information processing apparatus 2 can more accurately determine the road shape change by performing machine learning and then returning the normalized mesh to its original state and using it.

Next, the road shape change determination process by the map information processing apparatus 2 according to the present embodiment will be described with reference to FIGS. 7 and 8. When the map information processing apparatus 2 receives the instruction to start the road shape change process, the map information processing device 2 starts the determination process of step 19 from step 11 of FIG. In the following, the steps are indicated by S.

The map information processing device 2 executes parameter optimization based on the teacher data in advance before determining whether or not the road shape has changed. The road shape change determination unit 26 performs machine learning based on the teacher data (S11).

Information prepared in advance as teacher data is input to the road shape change determination unit 26. The information prepared in advance is used for setting a parameter for distinguishing from each other whether the difference data extracted in the past is an image showing a change in road shape or an image such as noise other than that. Then, the set parameter is held by, for example, the road shape change determination unit 26, or is stored in a memory accessible by the road shape change determination unit 26.

Next, the locus information acquisition unit 22 receives the probe data transmitted from each in-vehicle terminal 3 by the vehicle data reception unit 28. The trajectory information acquisition unit 22 stores the received probe data in the data storage unit 29. As a result, the track information acquisition unit 22 acquires the track information of each vehicle 4 included in the received probe data (S12).

The map information conversion unit 24 extracts map image information indicating the shape of the road from the information included in the map information of the vector data stored in the map information storage unit 21. The map information conversion unit 24 converts the map information from the three-dimensional vector data to the two-dimensional scalar data by extracting the map image information (S13).

In other words, the map information conversion unit 24 converts the vector format map image into the raster format map image. In the present embodiment, the map information conversion unit 24 converts the two-dimensional map information converted into scalar data into scalar data of a secondary mesh divided into secondary mesh units as mesh units of an arbitrary scale. The moving image drawing unit 23 creates probe data for a secondary mesh by normalizing the probe data acquired by the locus information acquisition unit 22 for a predetermined unit period.

The moving image drawing unit 23 draws a moving locus image of the secondary mesh based on the probe data of the normalized secondary mesh (S14).
The moving image drawing unit 23 accumulates the past probe data acquired by the locus information acquisition unit 22 for a specific period before creating the movement locus image based on the probe data acquired in a predetermined unit period. And normalize the data. The moving image drawing unit 23 creates probe data of the past normalized secondary mesh of data. Then, the moving image drawing unit 23 draws a moving locus image of the past secondary mesh based on the probe data of the normalized past secondary mesh. The movement trajectory image of the past secondary mesh is referred to for reference.

As the first step, the moving image drawing unit 23 draws a moving locus image of the secondary mesh drawn based on the probe data of the secondary mesh and a past secondary mesh drawn based on the probe data of the past secondary mesh. It is determined whether or not there is a difference by comparing with the moving locus image of (S15).

In the above-described embodiment, the moving image drawing unit 23 draws a moving locus image of the secondary mesh, but the present invention is not limited to this. The moving image drawing unit 23 can draw a moving locus image of a mesh of an arbitrary scale such as a primary mesh and a tertiary mesh. Further, the moving image drawing unit 23 is not limited to the case of comparing the moving locus image of the secondary mesh with the moving locus image of the past secondary mesh and determining whether or not there is a difference. For example, the moving image drawing unit 23 directly compares the latitude / longitude information included in the probe data of the secondary mesh with the latitude / longitude information included in the probe data of the past secondary mesh, and the difference is obtained. It is also possible to determine whether or not there is.

In the map information processing apparatus 2, if there is no difference between the movement locus image and the past movement locus image in S15, S16 to S19 shown in FIG. 7 are skipped, and the road shape change determination process is completed. That is, the map information processing apparatus 2 executes S16 to S17 only when there is a difference in the determination of S15. Therefore, the map information processing apparatus 2 can reduce the execution frequency of S17. Further, the map information processing apparatus 2 may set only the movement locus image having a difference from the movement locus image created based on the probe data accumulated in the past as the processing target of the difference data extraction processing by the difference extraction unit 25. can. Therefore, the map information processing apparatus 2 can reduce the processing load and improve the processing speed by reducing the processing targets in the difference extraction unit 25.

On the other hand, in the case of YES in S15 where there is a difference between the movement locus image and the past movement locus image, the difference extraction unit 25 is converted into the movement locus image drawn in S14 and scalar data in S13. Difference data representing the difference between the map information and the map image is extracted (S16).

The road shape change determination unit 26 determines whether or not the difference data newly extracted in S16 is a road shape change based on the result of machine learning learned in S11 (S17).
If the extracted difference data is YES in S18, which is a road shape change, the road shape change determination unit 26 outputs the extracted difference data as a road shape change (S19).

In this case, since the difference data indicates the change in the road shape, it is saved for use in the later map information creation process.
On the other hand, if the extracted difference data is NO in S18, which is not a road shape change, S19 is skipped and the road shape change determination process ends. In this case, the difference data is not information indicating a change in road shape, but is likely to be merely noise. Therefore, the extracted difference data may be discarded.

As described above, the map information processing apparatus 2 extracts only the map image information indicating the shape of the road from the map information of the vector data stored in the map information storage unit 21, and thereby maps the three-dimensional vector data. The information is converted into map information of two-dimensional scalar data having a simple data structure. As a result, in the map information processing apparatus 2, the movement locus image obtained from the probe data and the map information can be converted into the same simple two-dimensional data.

As a result, when the map information processing apparatus 2 detects a change in the road shape newly created by using machine learning, the processing target of the machine learning can be reduced to reduce the processing load. The map information processing apparatus 2 can easily determine the change in road shape by improving the processing speed and further improving the determination accuracy.

Next, the map information update process by the map information processing apparatus 2 according to the present embodiment will be described with reference to FIGS. 9, 10A, and 10B.
When the map information processing apparatus 2 receives the instruction to start the map information update process, the following map information update process is started. In the above-described embodiment, when the instruction to start the map information update process is received, the map information update process is started, but the present invention is not limited to this. For example, if a predetermined number of road shape changes are output in S19 of the road shape change determination process described above, the map information update process may be automatically started.

First, the map information updating unit 27 accumulates N difference data output in S19 of the road shape change determination process, and averages the error of the probe data (S31). Here, N is an arbitrary natural number. The probe data obtained by averaging the errors are shown in FIGS. 10A and 10B. FIG. 10B is an enlarged view of the region F1 of FIG. 10A. The map information updating unit 27 performs smooth processing on the difference data obtained by averaging the errors, that is, the data representing the road shape determined to be the road shape change (S32). The smoothed road shape corresponds to the shape of the newly created road.

The map information updating unit 27 acquires the map information of the mesh including the latitude and longitude of the road shape that has been smoothly processed in S32 among the map information of the vector data stored in the map information storage unit 21. Since the road shape determined based on the difference data is created based on the probe data, it has latitude / longitude information. Therefore, the map information updating unit 27 can acquire the map information of the corresponding mesh from the map information storage unit 21 by referring to the latitude / longitude of the probe data of the road shape.

As shown in FIGS. 10C and 10D, the map information updating unit 27 superimposes the two-dimensional road shape D2 smoothed in S32 on the road shape D3 included in the map information of the mesh of the acquired vector data (. S33). FIG. 10D is an enlarged view of the region F2 of FIG. 10C.

As shown in FIG. 10D, the map information updating unit 27 identifies the contact point between the two-dimensional road shape D2 that has undergone smooth processing in S32 and the road shape D3 included in the map information of the mesh of the vector data. Here, since the map information is represented by vector data, it includes information in the Z-axis direction. The Z-axis direction indicates a direction perpendicular to the ground. That is, the map information includes information representing the Z-axis coordinates of each road. The Z-axis coordinates are, for example, elevation. Therefore, the contact point between the road shape D2 generated from the difference data and the road in the map information is represented by the latitude, longitude, and altitude. Then, the map information updating unit 27 adds a node to each of the specified contacts (S34). In FIGS. 10D and 10E, it is exemplified as a node N1 and a node N2. Since the node added to the road shape D2 generated from the difference data includes information on the contact point representing the Z-axis direction of each road, the road shape D2 is converted into vector data.

The map information updating unit 27 uses the specified node to generate an axis of the road shape D2 of the vector data (S35).
The map information updating unit 27 corrects the road width of the road shape D2 obtained from the difference data so as to match the road width of the road shape D3 included in the map information. Then, the road shape D2 of the vector data in which the road width is corrected is connected to the road shape D3 included in the map information by a node (S36).

The map information updating unit 27 deletes the unnecessary road portion D4 and the added node from the road shape D3 included in the map information (S37). For example, the map information updating unit 27 deletes a portion where the probe data does not exist as an unnecessary road portion D4 between the nodes added to the original two-dimensional road shape D2 before conversion.

The map information updating unit 27 deletes the unnecessary road portion D4. The map information updating unit 27 updates the map information in the map information storage unit 21 based on the road shape D3 to which the road shape D2 of the vector data is connected (S38). When this process is completed, the map information update process is completed.

As a result, the two-dimensional difference data extracted as the road shape change can be returned to the three-dimensional vector data again to update the map information. As a result, the map information can be automatically formatted on a regular basis, and the time and effort required to update the map information can be reduced.

The map information processing device 2 of the present embodiment has a control unit, a temporary storage unit, a storage unit, and a wireless communication unit. The control unit executes the program read into the storage unit. The vehicle data receiving unit 28, the moving image drawing unit 23, the map information conversion unit 24, the difference extraction unit 25, the road shape change determination unit 26, and the map information update unit 27 of the present embodiment are used as control units of the map information processing device 2. included. The map information storage unit 21 and the data storage unit 29 of the present embodiment are included in the storage unit of the map information processing device 2. The temporary storage unit is a working area for developing programs and various data read from the storage unit. Each part such as a control unit, a temporary storage unit, a storage unit, and a wireless communication unit is connected to each other.

The map information processing device 2 of the present embodiment can be used, for example, as a technique for automatically formatting map data used in a car navigation system or an automatic driving support system.

The present invention is not limited to the above-described embodiment as it is, and the components can be modified and embodied within a range that does not deviate from the gist at the implementation stage. In addition, various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, all the components shown in the embodiments may be combined as appropriate. In addition, components across different embodiments may be combined as appropriate. It goes without saying that various modifications and applications are possible within the range that does not deviate from the purpose of the invention.

1 Map information processing system 2 Map information processing device 3 In-vehicle terminal 4 Vehicle 5 Communication terminal 21 Map information storage unit 22 Trajectory information acquisition unit 23 Moving image drawing unit 24 Map information conversion unit 25 Difference extraction unit 26 Road shape change determination unit 27 Map Information update unit 28 Vehicle data reception unit 29 Data storage unit

Claims (5)

A locus information acquisition unit that acquires locus information for each of multiple moving objects,
A moving image drawing unit that draws a moving locus image based on the locus information,
A map information storage unit that stores map information that has a map image showing the shape of the road,
A difference extraction unit that extracts difference data representing the difference between the movement locus image and the map image, and a difference extraction unit.
It is provided with a road shape change determination unit that determines whether or not the difference data is a road shape change by machine learning that uses the difference data extracted in the past as teacher data.
The moving image drawing unit is a map information processing system characterized in that the moving locus image is drawn based on the locus information normalized during a predetermined unit period. The map information processing system according to claim 1, wherein the moving image drawing unit collectively draws the moving locus images in a predetermined range. Further provided with a map information conversion unit that converts the map information from vector data to scalar data, including information indicating the position and shape of the road.
The map information conversion unit extracts only the map image information from the map information and converts it from vector data to scalar data.
The map information processing system according to claim 1 or 2, wherein the difference extraction unit extracts the difference data based on the map image of the map information converted into scalar data. Further provided with a map information update unit for updating the map information stored in the map information storage unit.
The map information is composed of mesh units separated into rectangles at regular latitude / longitude intervals.
The difference data determined to be a road shape change by the road shape change determination unit has the latitude / longitude information.
The map information update unit
The map information of the unit of the mesh including the latitude / longitude information of the difference data is acquired from the map information storage unit.
The first road shape, which is the map image shown by the difference data, and the second road shape, which is the map image included in the acquired map information, are superimposed to form the first road shape and the second road shape. Identify the point of contact with the road shape of
A node is added to the specified contact point, and the node includes the latitude, longitude, and altitude included in the map information.
Using the node, the first road shape of the vector data is generated.
The first road shape is modified to match the road width of the second road shape, and the second road shape is connected to the second road shape by the node.
By deleting unnecessary parts and the added node in the second road shape,
The map information processing system according to claim 3, which updates the map information. The road shape change determination unit performs the machine learning based on the mesh of the map information that has been normalized to the mesh in advance.
The map information processing system according to claim 4, wherein the map information updating unit updates the map information obtained by restoring the normalized mesh after the machine learning is performed.

Images