JP2025120386A - Information processing device - Google Patents

Abstract

The degree of congestion in a parking lot is estimated based on information obtained from vehicle travel.
The information processing device acquires location information, including parking locations, from mobile objects, and estimates parking lot information for parking lots located in off-road areas based on the location information of multiple mobile objects that have entered the off-road area. An estimation unit estimates the congestion level of the entire parking lot based on the parking locations.
[Selected figure] Figure 9

Description

The present invention relates to technology for collecting information about parking lots.

When distributing parking information, information on whether a parking lot is full or empty may be distributed. In conventional methods, vacancy information is generated by a management device installed in the parking lot that manages entrance and exit, based on information on the number of cars that can be parked, the number of cars entering, and the number of cars leaving. However, in parking lots that do not have such a management device, vacancy information cannot be generated, posing a problem in that information on the parking lot's congestion status, such as vacancy information, cannot be distributed.

Patent Document 1 describes a method for determining that a parking lot is full when a vehicle passing around the parking lot slows down by a predetermined threshold or more and passes through the entrance to the parking lot.

JP 2015-184820 A

However, the technology described in Patent Document 1 has issues with accuracy, as it is ultimately determined by the judgment of users passing by the parking lot. Furthermore, the above technology can only determine whether the parking lot is full or empty, and cannot estimate the parking lot's congestion level other than whether it is full or empty. Furthermore, there is no mention of the congestion level depending on the parking location within the parking lot.

The above is one example of the problem that the present invention aims to solve. The present invention aims to estimate the congestion level of a parking lot based on information obtained from vehicle movements.

The invention described in claim 1 is an information processing device comprising an acquisition unit that acquires location information, including parking locations, from mobile objects, and an estimation unit that estimates parking lot information regarding parking lots located in an off-road area based on location information of multiple mobile objects that have entered the off-road area, and the estimation unit estimates the overall congestion level of the parking lot based on the parking locations.

1 shows a configuration of an information processing system according to an embodiment. An example of probe data is shown. 10 shows an example of parking lot data. 10 is a flowchart of a parking lot area estimation process. 10 is a flowchart of a parking lot identification process and an estimation process. FIG. 10 is a diagram illustrating off-road driving. FIG. 10 is a diagram illustrating a method for estimating a parking lot area. An example of estimating a parking lot area from multiple driving planes is shown. 10 is a flowchart of a congestion level estimation process.

In one preferred embodiment of the present invention, an information processing device includes an acquisition unit that acquires location information, including parking locations, from mobile objects, and an estimation unit that estimates parking lot information regarding parking lots located in the off-road area based on the location information of multiple mobile objects that have entered the off-road area, and the estimation unit estimates the overall congestion level of the parking lot based on the parking locations.

The information processing device described above acquires location information, including parking locations, from mobile objects, and estimates parking lot information for parking lots located in off-road areas based on the location information of multiple mobile objects that have entered the off-road area. Here, the estimation unit estimates the congestion level of the entire parking lot based on the parking locations. This information processing device can estimate the congestion level of a parking lot by analyzing the parking locations of mobile objects in the off-road area.

In one aspect of the above information processing device, the estimation unit estimates the congestion level based on the concentration level of the parking location distribution. In this aspect, the congestion level is estimated based on the concentration level of the parking location distribution. Specifically, if the concentration level is high, the congestion level is estimated to be low, and if the concentration level is low, the congestion level is estimated to be high.

In another aspect of the above information processing device, the estimation unit calculates the degree of concentration based on the area of a standard deviation ellipse for the parking location. In a preferred example, the estimation unit divides the parking location into multiple areas and estimates the degree of congestion for each area by calculating the area of the standard deviation ellipse for each area. In another preferred example, the estimation unit estimates the degree of congestion for each of multiple areas based on parking locations throughout the parking lot.

In another aspect of the above information processing device, the acquisition unit acquires time information corresponding to the location information from the mobile object, and the estimation unit estimates the congestion level for each time period. In this aspect, the congestion level of the parking lot can be estimated for each time period.

In another preferred embodiment of the present invention, an information processing method executed by an information processing device includes an acquisition step of acquiring location information, including parking locations, from mobile objects, and an estimation step of estimating parking lot information regarding parking lots located in an off-road area based on location information of multiple mobile objects that have entered the off-road area, wherein the estimation step estimates the overall congestion level of the parking lot based on the parking locations. According to this information processing method, the congestion level in a parking lot can be estimated by analyzing the parking locations of mobile objects in the off-road area.

In another preferred embodiment of the present invention, a program executed by an information processing device equipped with a computer causes the computer to function as an acquisition unit that acquires location information, including parking locations, from mobile objects, and an estimation unit that estimates parking lot information regarding parking lots located in off-road areas based on location information of multiple mobile objects that have entered the off-road area, and the estimation unit estimates the overall congestion level of the parking lot based on the parking locations. By executing this program on a computer, the above-mentioned information processing device can be realized. This program can be stored and used on a storage medium.

Preferred embodiments of the present invention will now be described with reference to the drawings.
[System configuration]
Fig. 1(A) shows the configuration of an information processing system to which the present invention is applied. The information processing system includes a server 10 and a navigation device 20 mounted on a vehicle 3. The server 10 and the navigation device 20 are configured to be able to communicate wirelessly. Note that for the sake of convenience, only one vehicle 3 is shown in Fig. 1, but in reality, a large number of vehicles 3 communicate with the server 10.

Figure 1 (B) shows the configuration of the server 10. The server 10 includes a communication unit 11, a control unit 12, a map database (hereinafter, "database" will be referred to as "DB") 13, and a probe DB 14.

The communication unit 11 receives probe data from the navigation device 20. The probe data is generated by the navigation device 20 as the vehicle 3 travels. The map DB 13 stores map data. The map data includes link data indicating links corresponding to roads and node data corresponding to intersections, as well as parking lot data related to parking lots. The probe DB 14 stores probe data received from a large number of navigation devices 20.

The control unit 12 controls the entire server 10. The control unit 12 is configured with a computer such as a CPU, and performs predetermined processing by executing pre-prepared programs. Specifically, the control unit 12 performs processing such as receiving probe data from the navigation device 20 and estimating parking lot areas based on the probe data stored in the probe DB 14.

[Probe data]
Next, the probe data will be described. Fig. 2 shows an example of the probe data. The probe data is generated by the navigation device 20 at predetermined time intervals (e.g., every few seconds) as the vehicle 3 travels, and the example in Fig. 2 shows probe data at a plurality of locations generated at predetermined time intervals. As shown in the figure, the probe data includes "travel distance,""latitude,""longitude,""date and time,""speed,""direction of travel,""roadtype," and "back flag."

"Distance traveled" is the distance traveled by vehicle 3, and is the cumulative distance traveled measured by the odometer. "Latitude" and "Longitude" indicate the latitude and longitude of vehicle 3's location. "Date and time" indicates the date and time when each piece of probe data was generated. "Speed" indicates the speed of vehicle 3 at the time of measurement. "Direction of travel" indicates the direction of travel of vehicle 3 at the time of measurement. Note that the direction of travel is indicated by a direction with north being 0 degrees.

"Road type" is information indicating the type of road on which vehicle 3 is traveling, with expressways being assigned a value of "1," toll roads being assigned a value of "2," national highways being assigned a value of "3," prefectural roads being assigned a value of "4," and city roads being assigned a value of "5." In this embodiment, the navigation device 20 sets the road type to "-1" when vehicle 3 is traveling in an area other than a road. Therefore, when the road type is "-1," vehicle 3 is traveling off a road.

The "reverse flag" is a flag that indicates whether the vehicle 3 is backing up (moving backward). A reverse flag of "0" indicates forward movement, and a value of "1" indicates reverse (moving backward). The reverse flag is generated based on whether the gear of the vehicle 3 is in the reverse position.

[Parking lot data]
Next, parking lot data will be explained. Figure 3 shows an example of parking lot data. Parking lot data is prepared for each entrance of a parking lot. Therefore, if one parking lot has multiple entrances, multiple parking lot data will be prepared for one parking lot. As shown in the figure, the parking lot data includes "ID,""latitude,""longitude,""direction," and "name."

"ID" is an identification number that uniquely identifies the parking lot entrance. "Latitude" and "Longitude" are the latitude and longitude of the parking lot entrance. "Orientation" indicates the direction of travel when vehicle 3 enters the parking lot entrance. For example, if the parking lot entrance faces north, vehicle 3 will enter by traveling south from the north side of the parking lot, so the orientation of the entrance will be south. Note that in this example, orientation is indicated by an angle with north being 0 degrees.

"Name" is the name of the parking lot. Note that in the example of Figure 3, parking lot data is prepared for each parking lot entrance, but parking lot data may instead be prepared for each parking lot. However, even in this case, the parking lot data should include the latitude, longitude, and direction of the parking lot entrance. In other words, it should include information indicating the position and direction of each entrance to the parking lot (if there are multiple entrances).

[Parking lot area acquisition process]
Next, the parking lot area acquisition process will be described. The parking lot area acquisition process is executed by the server 10 and is a process for estimating the area of the parking lot based on probe data obtained from a large number of vehicles 3. This process is mainly executed by the control unit 12 of the server 10. In practice, this process is realized by a computer such as a CPU constituting the control unit 12 executing a program prepared in advance.

First, the server 10 receives probe data from a large number of vehicles 3 via wireless communication (step S10). The server 10 stores the received probe data in the probe DB 14.

Next, the server 10 performs parking lot identification (step S11). The parking lot identification process is a process of identifying parking lots using probe data and parking lot data in map data. Details of the parking lot identification process are shown in Figure 5 (A). First, the server 10 detects the start point of off-road travel based on the probe data (step S21).

Figure 6 is a diagram explaining off-road driving. As shown, multiple roads intersect, with roads running north-south and roads running east-west forming multiple plots (sites). A vehicle mark X indicates the location of vehicle 3 as indicated by probe data. A white vehicle mark X indicates a location on a prefectural road, and the probe data at this location includes the road type "4 (prefectural road)." A diagonally hatched vehicle mark X indicates a location on a city road, and the probe data at this location includes the road type "5 (city road)." A black vehicle mark X indicates a location off-road, and the probe data at this location includes the road type "-1 (off-road)."

As shown in the figure, assume that vehicle 3 is traveling west on road R3 (prefectural road), turns left at the intersection of roads R2 and R3 onto road R2 (city road), turns left again to enter parking lot section 50, drives through the parking lot, and finally parks vehicle 3 facing west (see dashed line 62). In this case, server 10 references the probe data showing the driving trajectory shown in Figure 6 and determines the point where the road type first becomes "-1 (off-road)" (the point indicated by dashed line 61 in Figure 6) as the start point of off-road driving.

Next, the server 10 references the parking lot data in the map DB 13 and searches for parking lots with entrances within a predetermined threshold range from the coordinates of the off-road traveling start point obtained in step S21 (step S22). From the multiple parking lots obtained by the search, the server 10 then identifies the parking lot whose entrance direction matches the traveling direction of the vehicle 3 at the off-road traveling start point and whose entrance is closest to the off-road traveling start point (step S23). Then, processing returns to the main routine shown in Figure 4.

Next, the server 10 determines whether the parking lot was identified in the parking lot identification process of step S11, i.e., whether a corresponding parking lot was found (step S12). If a corresponding parking lot was not found (step S12: No), the server 10 determines that the parking lot indicated by the currently obtained probe data has not yet been registered in the map DB 13, and registers the parking lot in the map DB 13 (step S13). Specifically, the server 10 generates new parking lot data and registers it in the map DB 13. At this time, the new parking lot data is assigned a new ID, the coordinates of the start point of the off-road traveling are set as "latitude" and "longitude," the direction of travel of the vehicle 3 at the start point of the off-road traveling are set as "heading," and the name of the facility or the like closest to the start point of the off-road traveling is set as "name."

In this way, when a new parking lot is registered in step S13, or when it is determined in step S12 that a parking lot has been identified, the server 10 extracts off-road driving data from the probe data and stores it in the probe DB 14 (step S14). In the example of Figure 2, data for points with a road type of "-1" is stored as off-road driving data.

Next, the server 10 performs an estimation process using the off-road driving data (step S15). Figure 5 (B) is a flowchart of the estimation process. First, the server 10 acquires off-road driving data obtained for the same parking lot from the probe DB 14 (step S31). Next, the server 10 generates a driving surface based on each off-road driving data (step S32). Here, a "driving surface" is a surface with a certain distance centered on the coordinates included in the off-road driving data. Now, assuming that the off-road driving data has the trajectory 42 shown in Figure 7 (A), the server 10 generates a driving surface 44 that has a width approximately the width of a standard vehicle relative to the trajectory 42, as schematically shown in Figure 7 (B). The server 10 generates such a driving surface 44 for each piece of off-road driving data.

The server 10 then spatially combines the multiple travel surfaces 44 generated in step S32 to estimate the parking lot area (step S33). Figure 8 shows an example of estimating a parking lot area from multiple travel surfaces 44, showing multiple travel surfaces 44 obtained from the same parking lot. In this example, it is possible to distinguish between an area where the vehicle 3 is moving in a relatively straight line and an area where the vehicle 3 has turned around and parked. Therefore, the server 10 estimates the area where the vehicle 3 has turned around and parked as the parking area 51, and the area where the vehicle 3 is moving in a relatively straight line as the travel area 52b. Note that in the example of Figure 8, two parking areas 51a and 51b are obtained, and two corresponding travel areas 52a and 52b are obtained, respectively.

In general, when parking a vehicle, the vehicle moves to the parking position at a certain speed, but when turning around to park the vehicle, the vehicle speed is sufficiently low. Therefore, the parking area 51 and the driving area 52 may be distinguished not only by the position coordinates of the vehicle 3 but also by taking into account the vehicle speed in the off-road driving data.

Furthermore, by combining one or more parking areas 51 and one or more driving areas 52 obtained in this way, the shape of the entire parking lot can be estimated. In the example of Figure 8, the overall shape of the parking lot is approximately the shape shown by the dashed line 48. Once the parking lot area has been estimated in this way, the parking lot area estimation process ends.

As described above, this embodiment makes it possible to estimate a parking lot area based on probe data from multiple vehicles, i.e., their driving history. In addition to the overall shape of the parking lot area, it is also possible to distinguish between driving areas and parking areas within the parking lot area.

[Estimation of congestion level]
Next, a method for estimating the congestion level of a parking lot based on the distribution of parking positions within the parking lot area is described. By dividing vehicle parking positions by time period, the distribution of parking positions for each time period can be obtained. Basically, when a parking lot is empty, drivers tend to park their vehicles in the most convenient areas within the parking lot. For example, in parking lots attached to facilities such as stores, parking spaces often fill up starting from the areas closest to the facility. On the other hand, when a parking lot is congested, the most convenient areas are often already filled, forcing drivers to park their vehicles in other areas, i.e., randomly occurring empty spaces. As a result, vehicle parking positions tend to be highly random and dispersed throughout the parking lot. Therefore, by analyzing the distribution of vehicle parking positions within the parking lot, if vehicle parking positions are concentrated in a certain area, it can be estimated that the parking lot is empty. On the other hand, if vehicle parking positions are dispersed throughout the parking lot, it can be estimated that the parking lot is congested.

Figure 9 is a flowchart of the process for analyzing congestion levels based on the vehicle's parking location. This process is performed by the server 10. First, the server 10 receives probe data from the navigation device 20 of the vehicle 3 (step S41).

Next, the server 10 obtains the coordinates of the parking positions of the vehicles 3 for each time period from the probe data (step S42). For example, the server 10 divides a day into three periods: morning, afternoon, and night, and obtains the coordinates of the parking positions of the vehicles for each time period. The server 10 then calculates the concentration of parking positions based on the obtained parking position coordinates, and estimates the congestion level for each time period (step S43). The process then ends. This makes it possible to obtain the congestion level of the parking lot for each time period.

Next, we will explain how to calculate the concentration of parking locations. The concentration of parking locations can be calculated using one of the following three methods:

(First Method)
The first method for calculating the concentration level utilizes a standard deviation ellipse. That is, the server 10 calculates a standard deviation ellipse based on the coordinates of the parking locations obtained in step S42 and calculates a predetermined percentage of the area of the standard deviation ellipse. For example, the server 10 calculates 90% of the area of the standard deviation ellipse. Basically, if parking locations are concentrated, the area of the standard deviation ellipse will be small, and if parking locations are dispersed, the area of the standard deviation ellipse will be large. Therefore, the server 10 compares the predetermined percentage of the area of the standard deviation ellipse with a predetermined threshold. If the percentage is smaller than the threshold, the server 10 estimates that the concentration level is high, i.e., the parking lot is empty. If the percentage is larger than the threshold, the server 10 estimates that the concentration level is low, i.e., the parking lot is crowded. In this way, the server 10 can estimate the congestion level for each time period based on the coordinates of the parking locations for each time period.

In the above example, one threshold was used to classify the congestion level of a parking lot into two levels: "empty" and "crowded." However, two or more thresholds may be used to classify the congestion level of a parking lot into three or more levels. For example, two thresholds may be used to classify the congestion level of a parking lot into three levels: "empty," "normal," and "crowded."

Note that the method of using the area of the standard deviation ellipse is a technique for calculating the concentration of parking positions in one location throughout an entire parking lot, and therefore cannot be applied as is when there are multiple concentration points throughout the entire parking lot. However, if the entire parking lot is divided into multiple parking areas and the area of the standard deviation ellipse is calculated for each parking area, this method can be applied even when parking positions are concentrated in multiple locations throughout the entire parking lot.

(Second Method)
The second method for calculating the concentration level uses the nearest neighbor distance method, which is a spatial analysis method. The nearest neighbor distance method classifies point distributions as either dispersed or concentrated, and calculates the "average nearest neighbor distance W" from the distribution of multiple points. Here, the average nearest neighbor distance W is the average value of the distance from each point to the nearest point, and is calculated using the following formula:

Here, as a classification criterion for each distribution, we consider the case where points are randomly distributed (following a uniform Poisson distribution) on a plane with area S. In this case, the expected value of the average nearest neighbor distance W is
is.

So,
This method can also be applied to cases where there are multiple concentration points in one parking lot.

(Third Method)
The third method for calculating the concentration level uses the K-function method, which is a spatial analysis method. The K-function method is a method for identifying distributions that cannot be distinguished using the nearest neighbor distance method, and calculates the K-function K(h) from the distribution of multiple points using the following formula:

Using the K-function, we can determine the scale (spatial range) at which points are concentrated or dispersed. If the points are randomly distributed, the expected value of the K-function is
is.

Therefore, on the scale of a circle of radius h,
This method can also be applied to cases where there are multiple concentration points in one parking lot.

In the above example, a day is divided into morning, noon, and night as time periods, but the application of the present invention is not limited to this. For example, a day may be divided into day and night, or into hourly intervals. A week may also be divided into seven days from Monday to Sunday. Furthermore, a year may be divided into months, or into four seasons. In this invention, the term "time period" is a concept that includes not only dividing a period into hours, but also division into days of the week or months.

In addition, in the above example, the congestion level is estimated for each time period, but the congestion level may also be estimated without limiting the time period.

3 Vehicle 10 Server 12 Control unit 13 Map DB
14 Probe DB
20 Navigation device

Claims (1)

an acquisition unit that acquires location information including a parking location from a mobile object;
an estimation unit that estimates parking lot information regarding parking lots present in the off-road area based on position information of a plurality of mobile objects that have entered the off-road area;
Equipped with
The estimation unit is an information processing device that estimates the overall congestion level of the parking lot based on the parking location.