JP7700752B2 - DATA GENERATION SYSTEM, DATA GENERATION METHOD, AND PROGRAM - Google Patents

Description

The present disclosure relates to a data generation system, a data generation method, and a program.

Patent Document 1 discloses a system that identifies facilities where a vehicle can be parked. A terminal device measures the current position of the vehicle and transmits probe information including the current position to an information providing server. The information providing server identifies a location where the vehicle is stopped for a predetermined period of time or longer as a facility where the vehicle can be parked.

JP 2009-169527 A

However, in Patent Document 1, there is a problem that in an environment where large and small parking lots are mixed, information about parking lots, etc. is generated erroneously.

The present disclosure has been made in consideration of the above background, and provides a data generation system, a data generation method, and a program that can appropriately generate data.

The data generation system according to this embodiment may include a parking position acquisition unit that acquires mobile body position information indicating the mobile body position of a mobile body, a facility information acquisition unit that acquires facility information including facilities and their position information, a clustering unit that clusters the mobile body positions to generate clusters according to the density of the mobile body positions, a range setting unit that sets a range of the mobile body based on the cluster, and a data generation unit that generates data in which the range is associated with the facility information.

In the above data generation system, the position acquisition unit may acquire parking position information indicating the parking position of a parked vehicle, the clustering unit may cluster the parking positions, the range setting unit may be a parking range setting unit that sets the parking range of the vehicle as the range of the vehicle, and the data generation unit may be a parking data generation unit that generates parking data in which the parking range is associated with the facility information.

In the above data generation system, the clustering unit may generate clusters based on a density distribution of the parking positions.

In any of the above data generation systems, the parking area may be set as an area where the density of parking locations is equal to or greater than a threshold value.

In any of the above data generation systems, the parking lot data generation unit may associate the parking lot range with the facility information based on the center of gravity of the cluster.

Any of the above data generation systems may further include a parking lot range database in which the parking lot ranges of known parking lots are registered, and the parking lot range database may be updated based on the parking lot data generated by the parking lot data generation unit.

In any one of the data generation systems described above, the facility information has attribute information including at least one of a size of the facility, a genre of the facility, and an available time of the facility;
The parking lot data generation unit may associate the parking lot range with the facility information based on the attribute information.

In any of the above data generation systems, the clustering unit may have a machine learning model that performs density-based clustering.

The data generation method according to this embodiment may include the steps of acquiring mobile object position information indicating the mobile object position of a mobile object, acquiring facility information including facilities and their position information, clustering the mobile object positions to generate clusters according to the density of the mobile object positions, setting a range of the mobile object based on the clusters, and generating data in which the range is associated with the facility information.

In the above data generation method, the mobile object position information may be parking position information indicating the parking position of a parked vehicle, the parking positions may be clustered, the parking area of the vehicle may be set as the range of the mobile object, and parking area data may be generated in which the parking area is associated with the facility information.

In the above data generation method, the clustering may generate clusters based on a density distribution of the parking positions.

In any of the above data generation methods, the parking area may be set as an area where the density of parking locations is equal to or greater than a threshold value.

In any of the above data generation methods, the parking lot range and the facility information may be associated based on the center of gravity of the cluster.

In any of the above data generation methods and any of the above programs, the parking range database may be configured to register parking ranges of known parking lots, and the parking range database may be updated based on the parking data.

In any of the above data generation methods, the facility information may have attribute information including at least one of the size of the facility, the genre of the facility, and the available hours, and the parking lot range may be associated with the facility information based on the attribute information.

In any of the above data generation methods, the clusters may be generated using a machine learning model that performs density-based clustering.

The program according to this embodiment causes a computer to execute the steps of acquiring mobile object position information indicating the mobile object position of a mobile object, acquiring facility information including facilities and their position information, clustering the mobile object positions to generate clusters according to the density of the mobile object positions, and setting a range of the mobile object based on the clusters and generating parking lot data in which the range is associated with the facility information.

In the above program, the mobile object position information may be parking position information indicating the parking position of a parked vehicle, the parking positions may be clustered, the parking area of the vehicle may be set as the range of the mobile object, and parking area data may be generated in which the parking area is associated with the facility information.

In the above program, the clustering may generate clusters based on a density distribution of the parking positions.

In any of the above programs, the parking area may be set as a parking lot area where the density of parking locations is equal to or greater than a threshold value.

In any of the above programs, the parking area range and the facility information may be associated based on the center of gravity of the cluster.

In any of the above programs, the parking range database may store the parking ranges of known parking lots, and the parking range database may be updated based on the parking data.

In any of the above programs, the facility information may have attribute information including at least one of the size of the facility, the genre of the facility, and the available hours, and the parking area may be associated with the facility information based on the attribute information.

In any of the above programs, the clusters may be generated using a machine learning model that performs density-based clustering.

The present disclosure provides a data generation system, a data generation method, and a program that can generate data appropriately.

1 is a block diagram showing a parking lot data generating system according to the first embodiment. FIG. 1 is a schematic diagram for explaining the facility and a parking lot. 4 is a flowchart showing a parking lot data generating method according to the first embodiment. 11 is a block diagram showing a parking lot data generating system according to a second embodiment of the present invention. 13 is a flowchart showing a parking lot data generating method according to the second embodiment. 13 is a block diagram showing a parking lot data generating system according to a third embodiment of the present invention. 13 is a flowchart showing a parking lot data generating method according to the third embodiment.

The present invention will be described below through embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential as means for solving the problems. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In each drawing, the same elements are given the same reference numerals, and duplicate explanations have been omitted as necessary.

First embodiment
1 is a block diagram showing a system 100 according to the embodiment 1. The system 100 includes a vehicle position DB (database) 1, a facility position DB 2, a parking lot data generating device 3, and a parking lot DB 4.

Vehicle position DB1 is a database that accumulates vehicle position information for multiple vehicles 7. The vehicle position information includes time series data on the vehicle position indicated by latitude, longitude, etc. For example, the vehicle position information associates the vehicle position with the measurement time of the vehicle position. Furthermore, the vehicle position information stores vehicle position information for each vehicle. For example, the vehicle position information associates the vehicle ID unique to the vehicle with the vehicle position. The vehicle position information may be acquired at regular time intervals for each vehicle.

For example, the vehicle 7 is provided with a probe 7a for measuring the vehicle position. For example, a positioning sensor of the Global Positioning System (GPS) can be used as the probe 7a. Alternatively, the vehicle position can be measured using a position estimation method using a wireless device such as WiFi (registered trademark), Bluetooth (registered trademark), or a beacon. In this case, the wireless device is used as the probe 7a. The vehicle position can be converted to latitude and longitude by combining the longitude and latitude of the wireless device and the vehicle position relative to the wireless device. In this case, the probe 7a does not have to be mounted on the vehicle. That is, the probe 7a may be mounted on a wireless device or the like. The probe 7a may also detect vehicle information indicating the vehicle speed, whether the engine is on or off, and the like. That is, the vehicle position DB1 may store vehicle information indicating the vehicle type and the vehicle state. The probe 7a transmits vehicle position information indicating the vehicle position to the parking lot data generation device 3 or the like.

The vehicle position information is not limited to flat parking lots, but may also include information specifying the vehicle's position in underground parking lots or multi-story parking lots. For example, the vehicle's altitude or the number of floors in the parking lot may be included in the vehicle position information. Furthermore, the frequency of obtaining the vehicle's position may include position information before and after parking, but may also include information while driving. If position information while driving is obtained, it can be used to correct the position when parking. Of course, positioning devices such as GPS are not limited to in-vehicle devices. For example, the vehicle position may be measured by a mobile terminal such as a smartphone carried by the driver or passenger.

The parking lot data generating device 3 may acquire vehicle position information from multiple vehicles 7 and store it in the vehicle position DB 1. Alternatively, a device other than the parking lot data generating device 3 may acquire vehicle position information and store it in the vehicle position DB 1.

The facility location DB2 is a database that stores facility location information indicating the location of a facility. For example, the facility location DB2 stores the location of each facility as a list. The facility location DB2 includes location information (coordinates) indicated by latitude, longitude, etc. for each facility. Here, examples of facilities include buildings and places with parking lots. The facility may also be a facility that does not include a building, such as a park. The facility may be a public facility or a private facility. The facility is not limited to a permanent facility, but may be a facility that is installed for a limited period of time. The facility is not limited to a parking lot such as a public parking lot or a private parking lot, but may include a house, a shop, an office, etc. The facility location DB2 stores information about facilities with parking lots.

The parking lot data generating device 3 generates parking lot data by referring to the vehicle position DB1 and the facility position DB2. In other words, the parking lot data generating device 3 generates parking lot data based on the vehicle position information and the facility position information. The parking lot data generating device 3 records the generated parking lot data in the parking lot DB4. Therefore, the parking lot DB4 stores parking lot data that links the facility with its parking lot.

The parking lot data generation device 3 will be described. The parking lot data generation device 3 includes a vehicle position acquisition unit 30, a parking lot position acquisition unit 31, a facility information acquisition unit 32, a clustering unit 33, a parking lot range setting unit 34, and a parking lot data generation unit 35.

The vehicle position acquisition unit 30 acquires vehicle position information from the vehicle position DB 1. Alternatively, the vehicle position acquisition unit 30 may acquire the vehicle position directly from the probe 7a. The user can also specify the period and area for which the vehicle position information is to be acquired. For example, the user can specify the area and period for which the vehicle position information is to be acquired. The vehicle position acquisition unit 30 can extract and read out vehicle position information for a specified period in a specified area.

The parking position acquisition unit 31 acquires parking position information indicating the parking position of the vehicle 7 based on the vehicle position information acquired by the vehicle position acquisition unit 30. For example, if the vehicle 7 is in the same position for a predetermined period of time or more, the parking position acquisition unit 31 sets that position as the parking position. The parking position information may include not only the parking position, but also information such as the parking time, parking date and time, and vehicle type.

Alternatively, the parking position acquisition unit 31 may acquire vehicle information from the vehicle 7 and set a parking position based on the vehicle information. The parking position acquisition unit 31 extracts the parking positions of multiple vehicles. This generates information indicating multiple parking positions on a map. In other words, a distribution diagram is generated in which a data distribution in which one parking position is one data point is shown on a map.

The vehicle information is, for example, information such as vehicle speed, engine on/off, and vehicle ID. Specifically, the parking position acquisition unit 31 can set the vehicle position when the engine is off as the parking position. Alternatively, the parking position acquisition unit 31 may set the vehicle position as the parking position when the speed is 0 km/h for a certain period of time or more.

The facility information acquisition unit 32 reads facility location information from the facility location DB 2. The facility location information is information that indicates the location of each facility.

The clustering unit 33 clusters the parking positions to generate clusters. The clustering unit 33 clusters the parking positions to generate clusters according to the density of the parking positions. For example, the parking positions are indicated by latitude and longitude coordinates. The clustering unit 33 performs clustering based on the density of the parking positions. The clustering unit 33 can perform clustering using a machine learning model generated by a machine learning method such as unsupervised learning. The machine learning model may also be generated using deep learning or the like. The clustering unit 33 may perform non-hierarchical clustering.

For example, the clustering unit 33 performs clustering based on the density of data points, using parking positions as data points. The clustering unit 33 detects an area where a dense area of data points is separated by a sparse area as one cluster. The clustering unit 33 can generate clusters based on the density distribution of parking positions. Areas with a high density of parking positions become clusters. For example, the clustering unit 33 has a machine learning model generated using an unsupervised machine learning clustering algorithm. The clustering unit 33 has a machine learning model that performs density-based clustering.

As density-based clustering, methods such as OPTICS (ordering points to identify the clustering structure) and DBSCAN (density-based spatial clustering of applications with noise) can be used. The clustering unit 33 determines the number of data points within a predetermined search distance (radius), and if the number of data points is equal to or greater than a predetermined value, expands the range of the cluster. The distance used for clustering can be Euclidean distance or the like. Of course, the clustering unit 33 may perform density-based clustering using other methods. For example, the clustering unit 33 can perform clustering using the mean-shift method. In this case, the clustering unit 33 determines the maximum point of density. Then, the clustering unit 33 sets the range including the maximum point and where the density is equal to or greater than the threshold as one cluster. In other words, positions where the density is less than the threshold are excluded from the cluster.

The parking lot range setting unit 34 sets the parking lot range based on the clusters. Each cluster corresponds to a parking lot, and the size of the cluster indicates the parking lot range. The parking lot range setting unit 34 sets the area defined by the cluster as the parking lot range. In other words, the boundaries of the clusters become the boundaries of the parking lot range.

Specifically, areas with dense data points become clusters that indicate areas where parking is possible, and areas with sparse data points indicate areas where parking is not possible. The parking lot range setting unit 34 sets dense areas where the density of data points is equal to or greater than a threshold as the parking lot range. The parking lot range setting unit 34 sets locations where the density of parking positions is equal to or greater than a threshold as the parking lot range. The parking lot range setting unit 34 sets an area defined by one cluster as one parking lot range. More specifically, the parking lot range setting unit 34 sets a continuous range where the density of data points is equal to or greater than a threshold as one parking lot range.

The parking lot data generation unit 35 generates parking lot data by associating the parking lot range with facility information. For example, the parking lot data generation unit 35 identifies the cluster closest to the facility. Then, the parking lot data generation unit 35 links the parking lot range indicated by the cluster closest to the facility with that facility. Since the parking lot data generation unit 35 can associate the parking lot range of the nearest parking lot with each facility, it is possible to generate parking lot data appropriately. For example, the parking lot data generation unit 35 can associate the parking lot range with facility information based on the center of gravity of the cluster. In other words, the parking lot data generation unit 35 extracts the parking lot range closest to the facility by taking the position of the center of gravity of the cluster as the position of the parking lot range.

Note that multiple parking lots may be provided separately for one facility. Therefore, the parking lot data generation unit 35 may associate multiple parking lot ranges with one facility. Alternatively, multiple facilities may share a parking lot. Multiple facilities may be associated with one parking lot range.

The process of generating parking lot data will be explained using Figure 2. Figure 2 is a plan view showing an example of three facilities A to C along road R. Facility A has a building BA and a parking lot PA. That is, the building BA of facility A is provided with a parking lot PA. Similarly, facility B has a building BB and a parking lot PB, and facility C has a building BC and a parking lot PC. The building BB of facility B is provided with a parking lot PB, and the building BC of facility C is provided with a parking lot PC.

When matching the nearest facility to each parking position, the parking lots of facilities A, B, and C are set as areas P1 to P3 divided by dividing line D. That is, facility A is matched with area P1 divided by dividing line D. Similarly, facility B is matched with area P2 divided by dividing line D, and facility C is matched with area P3 divided by dividing line D. Area P3 includes parking lot PA. Therefore, when matching the nearest facility to each parking position, there is a risk that the facilities and parking lots will not be matched appropriately. For example, area P3 corresponding to facility C includes parking lot PA and parking lot PC.

As such, if the method of this embodiment is not used, it will not be possible to set an appropriate parking area for each facility. In a situation where parking areas of different sizes are adjacent to each other, for example, it may happen that the center of a parking position in a large parking area is closer to the center of the surrounding parking areas than the center of the parking area itself. In such a situation, if the closest parking position is linked to a facility, an incorrect parking area will be calculated.

In this embodiment, the parking lot range setting unit 34 sets the parking lot range based on the cluster. Thus, parking lots PA to PC in FIG. 2 are set as separate parking lot ranges. The parking lot data generation unit 35 associates these parking lot ranges with the facilities. The parking lot data generation unit 35 can associate parking lots PA to PC with facilities A to C, respectively. This makes it possible to generate appropriate parking lot data.

For example, the clustering unit 33 determines that a group of parking positions with a certain density or higher is a cluster. The density of parking positions is low at the boundaries between parking lots, on roads R, and the like, and these are not determined to be within the parking area. Clusters are divided and generated for each parking lot. This makes it possible to associate the correct facilities with each parking lot, and prevents the parking area from deviating from the actual parking area.

In addition, the parking position extracted from the vehicle position information may contain noise points that are shifted from the accurate position due to various factors. For example, the parking position data may contain noise due to GPS errors when recording the vehicle position or the specifications of the collection device that serves as the probe 7a. Alternatively, noise may occur when data that was not anticipated when the parking position was extracted is present, such as when the vehicle 7 is stopped on the road due to a malfunction or the like. Such noise is likely to fall outside the area where the parking position has a certain density or higher. Parking positions where noise occurs have a low density and are not included in the cluster. Therefore, parking positions that are not included in the cluster can be regarded as noise and removed. This makes it possible to generate appropriate parking lot data.

Next, the parking lot data generation method according to this embodiment will be described with reference to FIG. 3. FIG. 3 is a flowchart showing the parking lot data generation method.

First, the vehicle position acquisition unit 30 acquires vehicle position information from the vehicle position DB1 (S101). For example, the vehicle position acquisition unit 30 reads vehicle position information in a region for a certain period from the vehicle position DB1. The user can specify the region for which parking lot data is to be generated. The user can also specify a period such as one month. The vehicle position acquisition unit 30 extracts vehicle position information for the specified region and the specified period. In addition, the vehicle position information may include vehicle information indicating the vehicle type, vehicle speed, engine on/off, etc.

The facility information acquisition unit 32 acquires facility locations from the facility location DB2 (S102). The facility information acquisition unit 32 may acquire facility locations in the designated area designated in S101.

The parking position acquisition unit 31 extracts the parking position from the vehicle position information (S103). For example, if the vehicle position remains within a certain range for a certain period of time or more, the parking position acquisition unit 31 can determine that position as the parking position. In this way, the parking position acquisition unit 31 can acquire parking position information indicating the parking position.

Furthermore, the parking position acquisition unit 31 may extract the parking position using the vehicle information. The parking position acquisition unit 31 can extract the parking position more accurately by using the vehicle speed and engine on/off information. That is, the parking position acquisition unit 31 may add extraction conditions according to the vehicle information. For example, the parking position acquisition unit 31 may extract the parking position using the position where the engine is switched on/off as an extraction condition. Alternatively, the parking position acquisition unit 31 may extract the parking position using a condition such as the vehicle speed being equal to or lower than a certain speed as an extraction condition. In other words, the parking position acquisition unit 31 may exclude from the parking positions positions where the engine cannot be switched on/off or positions where the vehicle speed is equal to or higher than a certain speed.

The clustering unit 33 clusters the parking positions to generate clusters according to the density of the parking positions (S104). The clustering unit 33 performs density-based clustering to generate appropriate clusters. Areas where the density is equal to or greater than a threshold are grouped together as a cluster. Areas where the density is equal to or less than a threshold are excluded from the cluster. For example, the density at a particular parking position can be calculated based on the number of parking positions within a certain distance from the particular parking position. Using a density-based clustering method, parking positions with a density equal to or greater than a certain density are connected from a particular parking position. The clustering unit 33 then treats the connected parking positions as a cluster.

The density threshold can be set taking into consideration regional characteristics and the number of vehicles contained in the probe data. The clustering unit 33 can use, for example, a density-based clustering method to find areas with a density above the threshold. Specifically, parking locations with a density above a certain level are connected from a certain parking location, and the connected parking locations are treated as a cluster. In this way, the clustering unit 33 can generate a cluster for each parking location.

The parking lot range setting unit 34 sets the parking lot range based on the cluster (S105). Since the area where the density is equal to or greater than the threshold is set as a cluster, the parking lot range setting unit 34 sets one cluster as one parking lot range. The parking lot range is position information that indicates the range of the parking lot where the vehicle 7 can be parked.

The parking lot data generation unit 35 links the parking lot range to an existing facility (S106). That is, the parking lot data generation unit 35 associates a parking lot range with each facility read from the facility location DB2. The parking lot data generation unit 35 generates parking lot data linking the facility with the parking lot range. The parking lot data generated by the parking lot data generation unit 35 is stored in the parking lot DB4. The parking lot data generation unit 35 links the facility location with the smallest distance to the center of gravity of the cluster. In accordance with regional characteristics, it is possible not to link the parking lot range to facilities that are more than a certain distance away.

Furthermore, the parking lot data generation unit 35 registers the parking lot that could not be associated with the facility as a new parking lot in the parking lot DB 4 (S107). In this way, parking lot data can be generated for parking lots that are not associated with existing facilities. Even when a new parking lot is established, parking lot data can be generated appropriately.

In this way, even in an area where parking lots of different sizes are adjacent to each other, the parking lot data generation device 3 can link the parking location to the correct parking lot. The parking lot data generation device 3 regards an area where the density of parking locations is equal to or higher than a certain level as the parking lot range. Therefore, it is possible to appropriately remove position information noise from GPS and data collection devices and errors due to insufficient accuracy of the facility location DB2 as noise. Therefore, it is possible to generate parking lot data with higher accuracy.
In addition, in the event of a disaster such as an earthquake, the use of parking lots may be restricted. Or, in order to restore the area, spaces other than existing parking lots may be used as new parking lots. Even in such cases, appropriate parking lot data can be generated quickly, which contributes to recovery from the disaster.

Embodiment 2
The parking lot data generating system and method according to the second embodiment will be described with reference to Fig. 4 and Fig. 5. Fig. 4 is a block diagram showing the system configuration. Fig. 5 is a flowchart showing the parking lot data generating method.

In this embodiment, a parking lot range DB5 is added to the system 100. The configuration other than the parking lot range DB5 is the same as in the first embodiment, so the description is omitted.

Parking lot range DB5 is a database that stores parking lot ranges of known parking lots. Parking lot ranges are position data that indicate parking areas in a parking lot where vehicles can be parked. Parking lot range DB5 stores data in which the parking lot range is associated with each parking lot. For example, the known parking lot ranges may be data that is automatically created by a computer, or may be data that is manually created by a user or administrator. Also, the parking lot ranges may be data that is partially created automatically and the rest created manually.

For example, a user or a computer detects the parking lot location based on a satellite image and sets the parking lot range. Alternatively, the parking lot range indicated in the parking lot data stored in the parking lot DB4 may be stored in the parking lot range DB5. In other words, the parking lot range of parking lot data generated in the past may be stored in the parking lot range DB5. The parking lot range in the parking lot range DB5 is updated with the latest information. Since the parking lot range DB5 pre-stores data indicating the parking lot range for some of the parking lot data, it can be used to determine the range of the parking lot to further improve accuracy.

In FIG. 5, step S206 has been added to the flowchart in FIG. 3. The processes other than step S206 are basically the same as those in embodiment 1, and therefore will not be described as appropriate. For example, steps S101 to S107 in FIG. 3 correspond to steps S201 to S205 and steps S207 to S208, respectively.

In step S205, the parking lot range setting unit 34 sets the parking lot range. Then, the parking lot data generation unit 35 sets the parking lot range newly set in step S205 as a range candidate. Then, the parking lot data generation unit 35 corrects the range candidate with the known parking lot range (S206). For example, if the known parking lot range and the range candidate overlap, the parking lot data generation unit 35 corrects both ranges as one parking lot range. If the known parking lot range and the range candidate partially overlap, the range obtained by merging the known parking lot range and the range candidate becomes the parking lot range. Also, if one range candidate overlaps with two known parking lot ranges, the parking lot data generation unit 35 may divide the range candidate by the two known parking lot ranges.

The parking lot data generation unit 35 links the known facilities to the corrected parking lot range (S207). Then, new parking lot data is registered based on the parking lot range that is not linked to the known facilities.

Furthermore, the parking lot data generation unit 35 may update the parking lot DB4 or the parking lot range DB5 based on the parking lot range corrected in S206. In this way, a highly accurate database can be generated based on the latest parking lot data and parking lot range. Also, while the parking lot range DB5 and the parking lot DB4 are shown as separate databases in FIG. 4, the parking lot range DB5 and the parking lot DB4 may be constructed as a common database.

Embodiment 3
The system and method according to the third embodiment will be described with reference to Fig. 6 and Fig. 7. Fig. 6 is a block diagram showing the system configuration. Fig. 7 is a flowchart showing the parking lot data generating method.

In this embodiment, a facility attribute DB6 is added to the system 100. The configuration other than the facility attribute DB6 is the same as in the first embodiment, so a description thereof will be omitted. The facility attribute DB6 stores attribute information indicating the attributes of the facility. For example, the attribute information may include data indicating the facility genre, such as whether the facility is a residential facility, a commercial facility, a public facility, or a private facility, or the attribute information may include data indicating the facility scale, such as the floor area and external area of the facility. Furthermore, the attribute information may include information such as the number of available users, the available period, the usage time zone, the parking time, the time zone during which parking is possible, the type of vehicle that can be parked, and the frequency of use per month.

In FIG. 7, the processing of steps S302 and S306 differs from the flowchart in FIG. 3. The processing other than steps S302 and S306 is basically the same as in embodiment 1, so the description will be omitted as appropriate. For example, steps S101, S103 to S105, and S107 in FIG. 3 correspond to steps S301, S303 to S305, and S307, respectively.

In step S302, the facility information acquisition unit 32 acquires facility information. In this embodiment, the facility information includes the facility location and attribute information. In other words, the facility information acquisition unit 32 reads out attribute information from the facility attribute DB 6 as facility information. It is preferable that the attribute information includes at least one of the size of the facility, the genre of the facility, and the available hours.

In step S306, the parking area is linked to the facility using attribute information. For example, the number of parked vehicles can be estimated according to the size of the facility. Thus, a parking area appropriate to the size of the facility can be linked. For example, if the nearest parking area is too large or too small for the facility size, the second closest parking area is assigned.

The parking lot data generation unit 35 can compare the attribute information with the parking lot range and associate the parking lot range with the facility based on the comparison result. The parking lot data generation unit 35 calculates an index indicating the degree of match based on the comparison result between the facility size and the area of the parking lot range. Then, the parking lot data generation unit 35 associates the parking lot range with the facility by considering the index according to the comparison result in addition to the distance between the facility and the parking lot range. Specifically, the parking lot data generation unit 35 appropriately weights the distance and the index, and assigns the parking lot with the highest degree of match to the facility.
Of course, the comparison between the attribute information and the parking lot range is not limited to the facility size. For example, the parking lot data generation unit 35 may make the comparison using the number of users, the vehicle type, the time period of use, the parking time, the frequency of use, etc. The parking lot data generation unit 35 makes the comparison using various attribute information. The parking lot data generation unit 35 calculates the degree of agreement by weighting and adding multiple indices according to the comparison results, and the parking lot. Furthermore, the degree of agreement may be indexed according to the regional characteristics.

Although the facility attribute DB6 is shown as a database separate from the facility location DB2, the facility attribute DB6 may be constructed as the same database as the facility location DB2. It is also possible to combine the second and third embodiments. In this case, the system 100 includes both the parking lot range DB5 and the facility attribute DB6. The parking lot data generation unit 35 can then use the known parking lot range to correct the parking lot range, and then associate the facility with the parking lot range.

The system 100 and the parking lot data generation device 3 may be realized by a single device, or may be distributed among multiple devices. For example, the system 100 may be a physically single device equipped with the parking lot data generation device 3 and each database. Alternatively, at least one database and the parking lot data generation device 3 may be mounted on physically different devices. For example, the parking lot data generation system and method may be realized by multiple information processing devices connected to a network performing distributed processing. For example, part of the processing of each functional block may be executed by the same device, and the rest may be executed by other devices. For example, the processing up to the extraction of the parking position and the processing after clustering may be executed by different devices. The parking lot data generation device 3 may also be called a parking lot data generation system.

In the above description, parking position information indicating the parking position of a vehicle is used as the mobile body position information indicating the position of the mobile body, but the mobile body position information may include information other than the parking position information of a vehicle. For example, the mobile body position information indicating the position of a mobile body may include a position where a mobile body such as a vehicle stays temporarily or a position where the mobile body is temporarily stopped. The mobile body position information may be a boarding and alighting position indicating a position where a mobile body moving at a low speed is boarded and alighted.

More specifically, for example, at a boarding/alighting area, passengers can board and alight from a shared vehicle or the like that is moving at a low speed. In other words, this embodiment can also be used for mobile bodies that can board and alight without completely stopping. System 100 may identify the range of such a boarding area based on the clustered mobile body positions. System 100 performs clustering based on mobile body positions including such boarding/alighting positions, parking positions, stopping positions, etc. The system sets a mobile body range indicating a boarding/alighting area, parking lot, stop, etc. based on the cluster. Then, the system associates the range of the mobile body with facility information. In this way, the system can generate data in which the range of a boarding/alighting area, parking lot, etc. is associated with a facility.

The system according to this embodiment includes a position acquisition unit that acquires mobile object position information indicating the mobile object position of a mobile object, a facility information acquisition unit that acquires facility information including facilities and their position information, a clustering unit that clusters the mobile object positions to generate clusters according to the density of the mobile object positions, a range setting unit that sets a range of the mobile object based on the clusters, and a data generation unit that generates data in which the range is associated with the facility information.

The method according to this embodiment includes the steps of acquiring mobile object position information indicating the mobile object position of a mobile object, acquiring facility information including facilities and their position information, clustering the mobile object positions to generate clusters according to the density of the mobile object positions, setting a range of the mobile object based on the clusters, and generating data in which the range is associated with the facility information.

In addition, a part or all of the above-mentioned generation method can be realized by a computer program. The program can be stored using various types of non-transitory computer readable media and supplied to a computer. Non-transitory computer readable media include various types of tangible storage media. Non-transitory computer readable media include, for example, magnetic recording media, magneto-optical recording media, CD-ROM (Read Only Memory), CD-R, CD-R/W, and semiconductor memory. Magnetic recording media include, for example, flexible disks, magnetic tapes, hard disk drives, etc. Magneto-optical recording media include, for example, magneto-optical disks, etc. Examples of semiconductor memories include mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, and RAM (Random Access Memory). The program may be supplied to the computer by various types of transient computer readable media. Examples of transient computer readable media include electrical signals, optical signals, and electromagnetic waves. The transient computer readable media can supply the program to the computer via wired communication paths such as electrical wires and optical fibers, or wireless communication paths.

100 System 1 Vehicle position DB
2 Facility location database
3 Parking lot data generating device 4 Parking lot DB
5. Parking Lot Area DB
6 Facility attribute DB
7 Vehicle 7a Probe 30 Vehicle position acquisition unit 31 Parking position acquisition unit 32 Facility information acquisition unit 33 Clustering unit 34 Parking lot range setting unit 35 Parking lot data generation unit

Claims (15)

a position acquisition unit that acquires moving object position information indicating moving object positions of a plurality of moving objects;
a facility information acquisition unit that acquires facility information including a facility to which a parking lot is attached and its location information;
a clustering unit that clusters the parking positions of the mobile object to generate clusters according to a density of the parking positions of the mobile object;
a parking lot area setting unit that sets a parking lot area for the moving object based on the cluster;
a parking lot data generating unit that generates parking lot data in which the facility information is associated with the parking lot range,
setting the moving body position of the moving body that has remained the same for a certain period of time or more as a parking position;
the clustering unit generates the clusters based on a density distribution of the parking positions,
A range in which the density of the data points indicating the parking positions is equal to or greater than a threshold and is continuous is set as a parking lot range;
the facility information has attribute information including at least one of a size of the facility, a genre of the facility, and an available time;
the parking lot data generation unit calculates a distance between a position of the facility indicated by the facility information and a center of gravity of the cluster;
comparing the attribute information with the parking lot area, and associating the parking lot area with the facility information according to a comparison result and the distance;
Data generation system. Extracting a parking area closest to the facility based on the distance;
The data generation system of claim 1, wherein when the size of the facility indicated by the attribute information is compared with the size of the nearest parking area range, if the nearest parking area range is too large or too small for the size of the facility, the second-closest parking area range is assigned to the facility. The position acquisition unit acquires parking position information including the parking position and a parking date and time,
The data generation system according to claim 1 or 2, wherein a comparison is made using the parking lot's operating time period to associate the parking lot range with the facility. A parking lot range database is further provided in which parking lot ranges of known parking lots are registered;
3. The data generating system according to claim 1, wherein the parking lot range database is updated based on the parking lot data generated by the parking lot data generating unit. The data generation system according to claim 1 or 2 , wherein the clustering unit has a machine learning model that performs density-based clustering. The computer
acquiring moving object position information indicating moving object positions of a plurality of moving objects;
acquiring facility information including a facility to which a parking lot is attached and its location information;
clustering the parking locations to generate clusters according to a density of parking locations for the mobile object location;
setting a parking area for the mobile object based on the cluster;
generating parking lot data in which the facility information is associated with the parking lot area;
setting the moving body position of the moving body that has remained the same for a certain period of time or more as a parking position;
generating the clusters based on a density distribution of the parking positions;
A range in which the density of the data points indicating the parking positions is equal to or greater than a threshold and is continuous is set as a parking lot range;
the facility information has attribute information including at least one of a size of the facility, a genre of the facility, and an available time;
calculating a distance between a position of the facility indicated by the facility information and a center of gravity of the cluster;
comparing the attribute information with the parking lot area, and associating the parking lot area with the facility information according to a comparison result and the distance;
Data generation method. Extracting a parking area closest to the facility based on the distance;
The data generation method according to claim 6, further comprising the steps of: comparing the size of the facility indicated by the attribute information with the size of the nearest parking area range; and if the nearest parking area range is too large or too small for the size of the facility, assigning the second-closest parking area range to the facility. Acquire parking location information including the parking location and the parking date and time,
The data generating method according to claim 6 or 7, wherein a comparison is made using the time periods when the parking lots are in operation, and the parking lot ranges are associated with the facilities. The parking range database contains the parking ranges of known parking lots.
The data generating method according to claim 6 or 7 , wherein the parking lot range database is updated based on the parking lot data. The data generating method according to claim 6 or 7 , wherein the clusters are generated using a machine learning model that performs density-based clustering. For computers,
acquiring moving object position information indicating moving object positions of a plurality of moving objects;
acquiring facility information including a facility to which a parking lot is attached and its location information;
clustering the parking locations to generate clusters according to a density of parking locations for the mobile object location;
setting a parking area for the mobile object based on the cluster;
generating parking lot data in which the facility information is associated with the parking lot area;
setting the moving body position of the moving body that has remained the same for a certain period of time or more as a parking position;
generating the clusters based on a density distribution of the parking positions;
A range in which the density of the data points indicating the parking positions is equal to or greater than a threshold and is continuous is set as a parking lot range;
the facility information has attribute information including at least one of a size of the facility, a genre of the facility, and an available time;
calculating a distance between a position of the facility indicated by the facility information and a center of gravity of the cluster;
comparing the attribute information with the parking lot area, and associating the parking lot area with the facility information according to a comparison result and the distance;
program. Extracting a parking area closest to the facility based on the distance;
The program according to claim 11, further comprising: comparing the size of the facility indicated by the attribute information with the size of the nearest parking area range, and if the nearest parking area range is too large or too small for the size of the facility, the program assigns the second-closest parking area range to the facility. Acquire parking location information including the parking location and the parking date and time,
The program according to claim 11 or 12, wherein a comparison is made using the time period during which the parking lot is used, and the parking lot range is associated with the facility. The parking range database contains the parking ranges of known parking lots.
The program according to claim 11 or 12 , wherein the parking lot range database is updated based on the parking lot data. The program according to claim 11 or 12 , wherein the clusters are generated using a machine learning model that performs density-based clustering.

Images