Lane formation determination device and method

Abstract

To analyze a human flow, especially to objectively evaluate whether a lane is formed by the human flow.SOLUTION: Provided is a lane formation determination device 10 comprising: an occupancy rate calculation unit 13 that sets a lane formation determination area, and calculates, using position information of a mobile entity including a human in the determination area, an occupancy rate that indicates the degree of occupancy of each of a plurality of attributes regarding movement in each cell constituting the determination area; an attribute determination unit 14 that identifies a cell attribute that represents each cell from among the plurality of attributes, in accordance with the occupancy rate; and a lane formation determination unit 15 that determines whether a lane is formed in the determination area, on the basis of the continuity of occupied cells the identified cell attribute of which are the same.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a technique for analyzing the movement status of people and the like, and to a technique for determining whether a lane is formed based on the movement status of moving objects such as people and the like. [Background technology]

[0002] In recent years, there has been a demand for analyzing human flow and other population movements for purposes such as controlling infectious diseases, evacuation during disasters, marketing, etc. In order to understand this human flow, people's trajectories are acquired.

[0003] For example, Patent Document 1 addresses the issue of "not simply calculating the degree of attention to an evaluation target, but also visually capturing the movement characteristics of people throughout the entire evaluation area." To this end, Patent Document 1 describes a speed distribution analysis device that includes an image data generation unit 120 that captures an image and generates image data from the image at predetermined time intervals; a person tracking unit 122 that tracks people within the image data; a divided area counting unit 124 that calculates the movement speed of the people from the displacement of the people measured by the person tracking unit and counts the movement speed of people located within each divided area obtained by dividing the entire area of ​​the image data into a grid; and a speed display unit 126 that displays the counted movement speed for each divided area. With this configuration, the speed distribution for each divided area can be visually displayed based on the movement information of people in the image data, making it easy to understand people's behavior relative to the target object. Patent Document 1 also discloses tracking the trajectory of people within the image data. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2006-85366 Summary of the Invention [Problem to be solved by the invention]

[0005] As described above, Patent Document 1 uses people's trajectories to analyze people flow. However, if an observer looks at the trajectories and makes a judgment to analyze people flow, the judgment will be qualitative, and there is a possibility that the evaluation results will be inaccurate. In particular, in people flow simulation, it is necessary to evaluate whether the simulation can be performed accurately, but if this evaluation depends on the observer, it is difficult to make an objective evaluation.

[0006] In view of the above circumstances, the present invention aims to analyze the movement situation of moving objects including people flow, and in particular to objectively evaluate whether a lane is formed by the movement of moving objects. [Means for solving the problem]

[0007] In order to solve the above problem, in the present invention, it is determined whether a lane has been formed based on the continuity (connection status) of cell attributes regarding the movement of a mobile object in each cell that constitutes a determination area.

[0008] More specifically, the lane formation determination device includes an occupancy calculation unit that sets a lane formation determination area and calculates, using position information of moving objects, including people, in the determination area, an occupancy rate calculation unit that calculates an occupancy rate indicating the degree of occupancy of each of multiple movement-related attributes in each cell that constitutes the determination area; an attribute determination unit that identifies a cell attribute that represents each cell from the multiple attributes in accordance with the occupancy rate; and a lane formation determination unit that determines whether a lane has been formed in the determination area based on the continuity of occupied cells having the same identified cell attribute. The present invention also includes a lane formation determination method using the lane formation determination device. Furthermore, the present invention also includes a program that causes the lane formation determination device to function as a computer and a storage medium for storing the program. [Effects of the Invention]

[0009] According to the present invention, it is possible to objectively evaluate whether a lane will be formed due to the flow of people. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram schematically showing the flow of people on a route 1 in one embodiment of the present invention. [Figure 2] 1 is a functional block diagram of a lane formation determination device according to an embodiment of the present invention; [Figure 3A] FIG. 10 is a diagram (part 1) for explaining an example of processing by a lane formation determination unit in one embodiment of the present invention. [Figure 3B] FIG. 10 is a diagram (part 2) for explaining an example of processing by the lane formation determination unit in one embodiment of the present invention. [Figure 3C] FIG. 10 is a diagram (part 3) for explaining an example of processing by the lane formation determination unit in one embodiment of the present invention. [Figure 4] FIG. 2 is a diagram showing layout data used in one embodiment of the present invention. [Figure 5] FIG. 10 is a diagram showing person position information used in one embodiment of the present invention. [Figure 6] FIG. 10 is a diagram showing attribute information used in one embodiment of the present invention. [Figure 7] 1 is a flowchart illustrating an overall process according to an embodiment of the present invention. [Figure 8] 10 is a flowchart showing details of steps S3 and S4 in one embodiment of the present invention. [Figure 9] 10 is a flowchart showing details of step S5 in one embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing the output contents of the lane formation determination result in one embodiment of the present invention. [Figure 11] 1 is a hardware configuration diagram showing an example of realizing a lane formation determination device according to a first embodiment. [Figure 12] FIG. 10 is a hardware configuration diagram showing an example of realizing a lane formation determination device according to a second embodiment. [Figure 13] FIG. 10 is a hardware configuration diagram showing an example of realizing a lane formation determination device according to a third embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0011] An embodiment of the present invention will be described below. In this embodiment, it is determined whether a lane is formed in a pedestrian flow based on pedestrian flow data relating to a pedestrian flow, which is one type of movement of a mobile object.

[0012] <Summary> First, an overview of this embodiment will be described. Fig. 1 is a diagram showing a schematic view of people flow on a route 1 in this embodiment. It also shows a schematic view of people flow data.

[0013] Figure 1(a) shows people moving towards their respective destinations on route 1, such as a road. In the figure, black circles indicate people moving to the right, and cross circles indicate people moving to the left. In other words, people are moving in the direction of the arrows in the figure. It can be seen that there are many people moving to the right at the top of route 1, and many people moving to the left at the bottom. As a result, as shown in Figure 1(b), lane 1, moving to the right, is formed at the top of route 1, and lane 2, moving to the left, is formed at the bottom.

[0014] In this way, a lane is a set of consecutive occupied cells formed on the route 1. An example of a lane can be expressed as a line of pedestrians with the same attributes. For example, when people heading left are visually seen lined up in a nearly straight line, it can be expressed as a line of people heading left, that is, a lane. Furthermore, in this embodiment, the determination area exemplified by the route 1 is the subject of lane formation determination. The determination area is then divided into multiple cells and managed.

[0015] <Configuration> Next, the configuration of this embodiment will be described. Fig. 2 is a functional block diagram of the lane formation determination device 10 in this embodiment. The lane formation determination device 10 has a function of determining whether a lane will be formed based on people flow data. Furthermore, simulation data or actual measurement data can be used as the people flow data. Furthermore, when simulation data is used, the lane formation determination device 10 may further have a simulator function for generating simulation data.

[0016] The lane formation determination device 10 has an input / output unit 11, a processing unit 12, and a storage unit 17. The input / output unit 11 inputs and outputs information with other devices, users, etc., and can be realized by a communication device, an input device, a display device, etc. The processing unit 12 performs various processes such as lane formation determination. To this end, the processing unit 12 has an occupancy rate calculation unit 13, an attribute determination unit 14, a lane formation determination unit 15, and a simulation unit 16.

[0017] First, the occupancy rate calculation unit 13 calculates an occupancy rate indicating the proportion of each person by attribute in each cell constituting the judgment lane from the people flow data. To do this, the occupancy rate calculation unit 13 counts up the number of cells each person enters by 1, and calculates the proportion of that count value as the occupancy rate. Note that the occupancy rate calculation unit 13 may also calculate the occupancy rate for each attribute based on the positional relationship between the cell each person entered last time and the cell each person entered this time.

[0018] Here, the attribute is information about the movement of a person, and can be the direction of movement, the speed of movement, the regularity of movement (randomness, degree of randomness), destination information (the direction of the destination in the movement), or a combination thereof. For this reason, the occupancy rate calculation unit 13 may store destination data in the person data for each time, and may regard the destination value data as the direction of movement, which is an example of an attribute.

[0019] Furthermore, the attribute determination unit 14 determines the attribute of each cell (hereinafter referred to as cell attribute) based on the difference in occupancy rate of each attribute of each cell, and the attribute of the cell that is occupied by the person with which attribute (majority attribute) each cell is occupied. For example, if the difference in occupancy rate is equal to or greater than a predetermined difference threshold (e.g., 0.5), the attribute determination unit 14 determines the attribute with the larger occupancy rate as the cell attribute of the cell. Furthermore, a cell whose cell attribute has been determined in this way is called an occupied cell.

[0020] Furthermore, the lane formation determination unit 15 determines whether a lane has been formed based on the continuity of occupied cells with the same cell attribute up to both ends of the determination area. The lane formation determination unit 15 may use various criteria to determine whether a lane has been formed. For example, (1) when at least one lane is formed, (2) when multiple lanes are formed, or (3) when multiple lanes with different cell attributes are generated. Furthermore, when checking the continuity of occupied cells with the same cell attribute, it may be acceptable for some cells to be disconnected. Examples of this will be described later.

[0021] Here, an example of processing performed by the lane formation determination unit 15 to check the continuity of occupied cells with the same cell attribute while extracting search candidate cells will be described with reference to Figs. 3A to 3C. Figs. 3A to 3C are diagrams for explaining an example of processing performed by the lane formation determination unit 15 in an embodiment. The processing proceeds in the order of Figs. 3A to 3C.

[0022] 3A, the lane formation determination unit 15 determines a group of cells at one end of a determination area 1713 as a start cell group 1711 from which a search candidate cell group starts. The lane formation determination unit 15 also determines a group of cells at the opposite end as a goal cell group 1712.

[0023] 3B, the lane formation determination unit 15 extracts one cell from the start cell group as a start cell 1711-1, which is the starting position of the search candidate cells. Note that when each cell in the start cell group is empty (does not indicate any attribute), the lane formation determination unit 15 determines that a lane will not be formed with that attribute.

[0024] Furthermore, the lane formation determination unit 15 determines whether the adjacent cell adjacent to the start cell (search candidate cell) has the same cell attribute. That is, the lane formation determination unit 15 extracts the adjacent cell in the direction indicated by the arrow in FIG. 3B and sets it as a new search candidate cell. Then, the lane formation determination unit 15 determines whether the new search candidate cell has the same cell attribute as the start cell (the attribute of the lane currently being determined). As a result, if the cell attributes are the same, the lane formation determination unit 15 extracts the adjacent cell that is the search candidate cell as an occupied cell.

[0025] Note that "adjacent" may be limited to adjacent in the direction of movement if the cell attribute of interest is the direction of movement, or may be limited to adjacent in a direction other than the opposite direction. That is, "adjacent" may mean adjacent in the direction from the start cell group to the goal cell group as shown by the arrows in Figure 3C, or adjacent in a diagonal direction or vertical direction (up and down in the figure).

[0026] Then, in FIG. 3C, the lane formation determination unit 15 starts from the identified occupied cell and sequentially determines whether adjacent cells will become new occupied cells. In FIG. 3C, in addition to the start cell, search candidate cells are indicated by bold frames. That is, cells "10," "11," "21," "22," "32," "42," and "41" are consecutive occupied cells with the same cell attribute. Also, in the figure, cells with dots outside the bold frame (such as "20" and "12") indicate cells that are determined not to have the same cell attribute. These include occupied cells with different cell attributes and blank cells with no identified cell attribute.

[0027] In this way, the lane formation determination unit 15 determines whether or not a lane is formed based on whether or not the same cell attributes are consecutively adjacent from the start cell 1711-1 to the goal cell in the goal cell group 1712. In this way, the lane formation determination unit 15 propagates and determines the cell attributes of each cell one after another, and if the new occupied cell is a cell included in the goal cell group 1712, it determines that a lane related to the cell attribute has been formed, and temporarily ends this process.

[0028] Furthermore, the lane formation determination unit 15 performs the above search for each cell attribute. As a result, if the above-mentioned conditions are satisfied, the lane formation determination unit 15 determines that a lane has been formed in the determination area 1713. For example, when the movement direction is used as the cell attribute, if lanes are formed in both directions (left and right), it is determined that a lane has been formed in the counterflow.

[0029] The lane formation determination unit 15 may have a function to evaluate the simulator that executed the simulation based on its own determination result. The lane formation determination unit 15 is expected to evaluate the simulator as passing when it determines that lanes have been formed, or to calculate an evaluation score according to the number of lanes that have been formed. This evaluation function is optional and can be omitted. Furthermore, this function may be provided in a device separate from the lane formation determination device 10.

[0030] The simulation unit 16 also executes a simulation of people flow and generates simulation data. The simulation unit 16 is an optional component and is not essential. In particular, when actual measurement data is used, the simulation unit 16 can be omitted. The simulation unit 16 may also be provided in a device separate from the lane formation determination device 10. In this embodiment, lane formation can be determined from at least one of the simulation data and the actual measurement data.

[0031] The storage unit 17 also stores various information and data used to determine lane formation. That is, the storage unit 17 stores layout data 171, person position information 172, attribute information 173, simulation data 174, and actual measurement data 175. These various information and data will be described below in <Information and Data>.

[0032] The lane formation determination device 10 described above may be implemented by a computer as long as it is capable of executing the above-described functions. Examples of implementation by a computer will be described in the following embodiments.

[0033] <Information and Data> Various pieces of information and data used in this embodiment will be described below. First, FIG. 4 is a diagram showing layout data 171 used in this embodiment. The layout data 171 is data that holds structural information representing a predetermined area, such as route 1. That is, as shown in FIG. 4, the layout data 171 indicates that the predetermined area is composed of multiple cells (00 to 94). Furthermore, the layout data 171 contains the positional relationship of the cells in the lane formation determination described above, as well as their adjacency relationship. That is, the layout data 171 contains information for converting cell positions, such as cell size, offset relationship, and cell shape. Note that a cell in this embodiment is a unit space that can be assigned attributes and managed. For example, a two-dimensional space can be divided into squares, such as 80 cm squares, and each square can be treated as a cell, as it is a unit for managing information such as the number of people. Furthermore, the shape of the cells can be managed as a square or other shape, such as a hexagon.

[0034] FIG. 5 is a diagram illustrating the person position information 172 used in this embodiment. The person position information 172 is sequence information indicating the position of a person in a people flow. In other words, it indicates the position information of each person at each time. For this reason, as shown in FIG. 5, the person position information 172 records, for each person ID, information identifying a cell (the number in the layout data 171) as the position information at each time. In other words, person ID = "1" moves to cell "01" at time 00:00.1, to cell "11" at time 00:00.5, and to cell "21" at time 00:01.0. Note that the person ID in this embodiment may be information that can be used to identify a person from other people within a specified area; it does not necessarily have to be information that can identify an individual by name or the like. Furthermore, the person position information 172 can be created from people flow data (simulation data 174 or actual measurement data 175). Note that the person position information 172 is an example of the position information of a moving object.

[0035] 6 is a diagram showing attribute information 173 used in this embodiment. The attribute information is information indicating the cell attribute representing each cell. In this embodiment, the occupancy rate and the difference therebetween for calculating the cell attribute are also recorded. These may be managed as separate information.

[0036] As shown in FIG. 6, the attribute information 173 records the occupancy rate (for each attribute), the difference, and the cell attribute for each cell. Here, a cell is information that identifies a cell that constitutes a specified area, and can be realized by a number in the layout data 171. The occupancy rate indicates the proportion of each attribute in the cell. In this embodiment, left, right, top, bottom, and stop are used as attributes to indicate the direction of movement. Left, right, top, and bottom indicate the directions in FIG. 4, and stop indicates that the cell is stopped. These are just examples of attributes, and some of the attributes such as stop may be omitted, or other attributes may be added.

[0037] The difference indicates the difference between the maximum occupancy rate and the second-largest value (second value) for the cell. For example, for cell "00," the right occupancy rate "0.65" minus the left occupancy rate "0.25" = difference "0.4." For cell "01," the right occupancy rate "0.8" minus the top occupancy rate "0.15" = difference "0.65."

[0038] If this difference is greater than a preset difference threshold, the attribute with the largest occupancy rate becomes the cell attribute. For example, for cell "00," the difference is "0.4," which is less than the difference threshold of "0.5," so the cell attribute becomes "-" (blank cell). For cell "01," the difference is "0.65," so the cell attribute becomes "right." Similarly, the cell attribute of cell "02" also becomes "right." Cells whose cell attributes have been identified in this way are referred to as occupied cells.

[0039] The simulation data 174 and the actual measurement data 175 are data indicating the movement status of each person, and are data as shown in Fig. 1. The actual measurement data can be realized by image data captured by a camera or data processed from the image data. This concludes the explanation of the information and data used in this embodiment.

[0040] <Processing flow> Next, the processing flow of this embodiment will be described. First, Fig. 7 is a flowchart showing the overall processing of this embodiment. In step S1 of Fig. 7, the input / output unit 11 acquires people flow data. As described above, simulation data 174 and actual measurement data 175 are used for the people flow data. This step may include performing a simulation in the simulation unit 16 and taking photographs to acquire the actual measurement data 175.

[0041] In step S2, the occupancy rate calculation unit 13 sets a judgment area 1713 to be judged. To this end, the input / output unit 11 may accept specification of both ends of the judgment area 1713 (start cell group 1711 and goal cell group 1712) from the user, or the occupancy rate calculation unit 13 may specify both ends of the judgment area 1713 according to a predetermined setting rule.

[0042] Furthermore, in step S3, occupancy rate calculation unit 13 calculates the occupancy rate of each cell in judgment area 1713. Then, in S4, attribute determination unit 14 determines the cell attribute of each cell in judgment area 1713 using the occupancy rate calculated in step S3. These steps S3 and S4 execute the processing described in <Configuration>, the details of which will be described using FIG. 8. FIG. 8 is a flowchart showing the details of steps S3 and S4 in this embodiment. In FIG. 8, steps S31 and S32 show the details of step S3, and steps S41 to S45 show the details of step S4.

[0043] First, in step S31, the occupancy rate calculation unit 13 extracts cells from the determination area 1713. This does not matter as long as each cell from the determination area 1713 can be extracted one by one, and the order of extraction does not matter. Then, in step S32, the occupancy rate calculation unit 13 calculates the occupancy rate for each attribute for each extracted cell. To this end, the occupancy rate calculation unit 13 uses the person position information 172 to calculate the occupancy rate of the attribute information 173 shown in FIG. 7.

[0044] In step S41, the attribute determination unit 14 calculates the difference in occupancy rate for each cell. That is, the attribute determination unit 14 calculates the maximum occupancy rate minus the second maximum occupancy rate. In step S42, the attribute determination unit 14 compares the calculated difference with a difference threshold (e.g., 0.5). As a result, if the difference is greater (Y), the process proceeds to step S43. In addition, if the difference is equal to or less than the difference threshold (N), the process proceeds to step S44.

[0045] In step S43, the attribute determination unit 14 determines the attribute of the maximum value as the cell attribute of the corresponding cell. Then, the attribute determination unit 14 registers this cell attribute in the attribute information 173. In step S44, the attribute determination unit 14 determines that the corresponding cell is unoccupied, that is, a blank cell. Then, the attribute determination unit 14 registers in the attribute information 173 that the cell is a blank cell (-).

[0046] In step S45, the attribute determination unit 14 determines whether processing has been completed for each cell in the determination area 1713. If the processing has been completed, the process proceeds to step S5. If the processing has not been completed (N), the process proceeds to step S31, where the occupancy rate calculation unit 13 extracts the next cell.

[0047] This concludes the description of steps S3 and S4. However, the following processing may be performed in step S3. That is, the occupancy calculation unit 13 may be configured to calculate the occupancy only within a period in which the sum of the proportions of different attributes within the determination area 1713 is equal to or greater than a certain value. Specifically, the occupancy calculation unit 13 calculates the proportion of the number of people within the determination area 1713 to the number of targets included in the people flow data. That is, the occupancy calculation unit 13 calculates the sum of the proportions of the number of people with different attributes. Then, the occupancy calculation unit 13 sets the time period (before and after) in which this sum is equal to or greater than a certain value (or a maximum) as the evaluation period, and calculates the occupancy from the person position information 172 within the evaluation period. By processing in this manner, the time period after people start to leave the determination area 1713 can be excluded from the determination. For example, determinations after the time period in which people move in opposite directions are excluded, making it possible to narrow down the determination to a more meaningful time period.

[0048] This concludes the explanation of Fig. 8, and returning to Fig. 7, we will continue to explain the processing flow of this embodiment. In step S5, the lane formation determination unit 15 uses the cell attributes determined in step S4, that is, the cell attributes of the attribute information 173, to determine whether a lane has been formed in the determination area 1713. The outline of this processing is as explained in <Configuration>, but more detailed processing content will be explained using Fig. 9. Fig. 9 is a flowchart showing the details of step S5 in this embodiment.

[0049] In step S501, the lane formation determination unit 15 identifies the start cell group 1711 and the goal cell group 1712 at which the search starts in the determination area 1713 set in step S2. At this time, it is desirable for the lane formation determination unit 15 to identify the number and types of lane types (attributes) to be determined, but this may be performed at other times. For this purpose, the input / output unit 11 may accept instructions from the user, or these may be stored in the storage unit 17 in advance.

[0050] In step S502, the lane formation determining unit 15 extracts a start cell 1711-1, which is the starting position of the search, from the start cell group 1711 and adds it to the search candidate cell group. There is no restriction on the rule for extracting the start cell 1711-1.

[0051] In step S503, the lane formation determination unit 15 determines the cell attribute of the extracted start cell 1711-1. To do this, the lane formation determination unit 15 uses the results of steps S43 and S44. That is, the cell attribute of the attribute information 173 is used.

[0052] Furthermore, in step S504, the lane formation determination unit 15 extracts one search candidate cell from the group of search candidate cells and extracts its adjacent cells. This extraction can be performed using the method using the above-mentioned FIGS. 3A to 3C. Furthermore, in step S504, multiple cells can be extracted as adjacent cells. Then, in step S505, the lane formation determination unit 15 identifies the cell attributes of the extracted adjacent cells. This identification can be performed in the same way as in step S503. Note that the processing order of steps S503 and S505 does not matter.

[0053] Furthermore, in step S506, the lane formation determination unit 15 compares the cell attributes of the neighboring cells of the search candidate cell. As a result, if there is a match (Y), the lane formation determination unit 15 determines that the neighboring cell is a search candidate cell and adds it to the search candidate cell group. Then, the process proceeds to step S508. If there is a mismatch (N), the lane formation determination unit 15 determines that the neighboring cell is not a search candidate cell and the process proceeds to step S507. Note that if there are multiple neighboring cells, the determination is "Y" if one or more neighboring cells match, and "N" otherwise. If the determination is "Y", multiple neighboring cells with matching attributes may be added to the search candidate cell group.

[0054] Furthermore, in step S507, the lane formation determination unit 15 determines that a lane including the start cell 1711-1 extracted in step S402 will not be formed. Furthermore, in step S508, the lane formation determination unit 15 determines whether the adjacent cell is included in the goal cell group 1712. Here, the lane formation determination unit 15 performs this determination for adjacent cells with matching attributes. As a result, if one or more adjacent cells are included in the goal cell group 1712 (Y), the process proceeds to step S509. If not (N), the process proceeds to step S504. Then, the lane formation determination unit 15 extracts one search candidate cell from the search candidate cell group and extracts its adjacent cells, thereby executing the process of step S504.

[0055] In step S509, the lane formation determining unit 15 determines that one lane is formed in the start cell 1711-1, that is, the attribute indicated by the cell attribute determined to match in step S506.

[0056] Furthermore, in step S510, the lane formation determination unit 15 checks whether a lane has been determined to be formed for an attribute (another attribute) other than the attribute determined in step S509 in the corresponding determination area. That is, it checks whether a lane has been formed for all types (attributes) to be determined in step S501. As a result, if it is determined that lanes for other attributes have been formed, that is, if it is determined that lanes have been formed for all types (attributes) to be determined (Y), the process proceeds to step S511. Furthermore, if there is an attribute for which a lane has not been determined to be formed (N), the process proceeds to step S512.

[0057] Furthermore, in step S511, the lane formation determination unit 15 determines that a lane has been formed in the corresponding determination area. Then, this process ends. Furthermore, in step S512, the lane formation determination unit 15 checks whether this process has been executed for each start cell in the start cell group 1711. That is, it checks whether the lane formation determination unit 15 has made a determination regarding the corresponding determination. As a result, if there are no unprocessed start cells remaining and processing for each cell group in the start cell group 1711 has been completed (Y), the process proceeds to step S6. Furthermore, if there are unprocessed start cells remaining, the process proceeds to step S502, where the lane formation determination unit 15 extracts another start cell from the start cell group 1711. Note that here, it is desirable that the lane formation determination unit 15 extracts a start cell with a cell attribute different from the lane formation attribute as the other start cell.

[0058] The lane formation determination unit 15 outputs these determination results via the input / output unit 11. FIG. 10 is a diagram showing the output contents of the lane formation determination results in this embodiment. In FIG. 10, a display screen 1110, which is an example of the input / output unit 11, includes an area display area 1111 and a comment display area 1112. The area display area 1111 displays the formed lanes 1111-1 and 1111-2 in a determination area separated by cells. The lanes 1111-1 and 1111-2 may be distinguished by adding arrows indicating their shapes or attributes, as shown in the figure, or by displaying them in different colors. Furthermore, the occupancy rate of each attribute may be written in each cell. Furthermore, the area display area 1111 may be limited to the determination area. Furthermore, in addition to the lanes, blank cells may be displayed in a manner that distinguishes them from other cells. Furthermore, the comment display area 1112 displays the lane formation determination results and an evaluation of the simulation. These are not essential and may be omitted.

[0059] In the process of step S5 shown in FIG. 9, it is determined that lanes of a plurality of mutually different attributes have been formed within the determination area. In other words, if there is an attribute for which it is determined that a lane is not formed, it is determined that the lane is not restricted in the determination area. Furthermore, if it is determined that lanes of a plurality of attributes have been formed, the subsequent process is omitted. This makes it possible to reduce unnecessary searches (determination of lane formation). Note that these are merely examples, and if lanes are formed with one or more attributes, it may be determined that a lane is formed in the determination area, or it may be determined that multiple lanes are formed for the same attribute.

[0060] In this embodiment, it is determined that one lane is formed when cells with the same cell attribute are continuous without any missing cells. However, it may also be determined that a lane is formed even if some cells are not connected. The following describes this process. If the lane formation determination unit 15 determines in step S506 that the attributes do not match, it checks whether the determination of mismatch has continued. Furthermore, the lane formation determination unit 15 checks whether the number of mismatched attributes in a series of searches using an arbitrary start cell is equal to or greater than a certain number or a certain percentage depending on the determination area. If these results show that there are no consecutive mismatch determinations and the number of attributes is equal to or less than a certain number or a certain percentage, the process proceeds to step S508, where the search continues. As a result, it is possible to determine that one lane is formed even if some cells are missing.

[0061] Furthermore, when the lane formation determination unit 15 uses the simulation data 174, the lane formation determination unit 15 may evaluate the simulation data 174, the simulator that generated the simulation data, or the simulation process, depending on the results of step S511 and step S512. This is as explained in the above <Configuration>.

[0062] This concludes the explanation of FIG. 9. Now, returning to FIG. 7, we will continue to explain the processing flow of this embodiment. In step S6, the lane formation judgment unit 15 judges whether it is necessary to reset the judgment area. This may be determined in response to a user instruction from the input / output unit 11, or may be determined according to conditions previously set in the storage unit 17. These conditions may include whether the size (number of cells) of the judged judgment area is equal to or smaller than a predetermined value, the number of judgments, etc. As a result, if it is determined that resetting is necessary (Y), the process proceeds to step S7.

[0063] In step S7, the judgment area is reset so as to reduce the judgment area set in step S2. In other words, the reset judgment area is included in the judgment area set in step S2. Therefore, the cell attributes of the cells that make up the reset judgment area have already been determined. Therefore, step S5 is executed after step S7. Note that the reset area does not have to be included in the judgment area. In this case, steps S3 and S4 are executed for each cell whose cell attributes have not been determined.

[0064] If resetting is not necessary (N), this processing flow ends. Note that steps S6 and S7 may be omitted. This concludes the description of the processing flow of this embodiment, and below, we will describe each example showing a specific implementation example of this embodiment. [Example]

[0065] Example 1 is an example in which lane formation determination is performed using simulation data 174. FIG. 11 is a hardware configuration diagram showing an example of implementing a lane formation determination device 10 in Example 1. In FIG. 11, the lane formation determination device 10 is connected to a file server 100 via a network 40. Here, the lane formation determination device 10 can be implemented by a computer, and includes a processing device 101, an input device 102, a display device 103, a communication device 104, a memory 105, and a secondary storage device 106, which are connected to each other via a communication path.

[0066] First, the processing device 101 can be realized by a processor such as a CPU, and performs calculations in accordance with a lane formation determination program 110 and a simulation program 120 stored in a secondary storage device 106. Therefore, the processing device 101 corresponds to the processing unit 12 in FIG.

[0067] The input device 102 has a function of receiving instructions and operations from a user, and can be realized by an input device such as a mouse or a keyboard. The display device 103 has a function of displaying the lane formation determination results, simulation data 174, actual measurement data 175, etc., shown in Fig. 10. The display device 103 can be realized by a monitor or the like. The input device 102 and the display device 103 correspond to the input / output unit 11 in Fig. 2, and may be realized by a single device such as a touch panel.

[0068] 2. The communication device 104 has a function for connecting to the file server 100 via the network 40, and transmits and receives various information and data to and from the file server 100. The communication device 104 also corresponds to the input / output unit 11 in FIG.

[0069] 2. The memory 105 and the secondary storage device 106 store the lane formation determination program 110, the simulation program 120, and information used for processing by the processing device 101. The secondary storage device 106 can be implemented as a so-called storage device, and stores the lane formation determination program 110 and the simulation program 120. The secondary storage device 106 can be implemented as various storage media, such as an external hard disk drive (HDD), solid state drive (SSD), or memory card.

[0070] The lane formation determination program 110 is composed of an occupancy rate calculation module 111, an attribute determination module 112, and a lane formation determination module 113. Each of these modules may be realized as an independent program. Each of these modules is configured to execute the function of each unit shown in FIG. 2. The correspondence between each of these modules and programs and each unit in FIG. 2 is as follows: Occupancy rate calculation module 111: Occupancy rate calculation unit 13 Attribute determination module 112: attribute determination unit 14 Lane formation determination module 113: Lane formation determination unit 15 Simulation Program 120: Simulation Section 16 Furthermore, the processing device 101 executes a simulation of human movement in accordance with the simulation program 120 and generates simulation data 174.

[0071] Furthermore, the file server 100 stores layout data 171, person position information 172, attribute information 173, and simulation data 174. The lane formation determination device 10 can acquire this information from the file server 100 and determine lane formation. At least a part of this information may be stored in the secondary storage device 106. When the layout data 171, person position information 172, attribute information 173, and simulation data 174 are stored in the secondary storage device 106, connection to the file server 100 and the network 40 can be omitted. In other words, the lane formation determination device 10 can be realized as a so-called standalone device. This concludes the description of the first embodiment. [Example]

[0072] The second embodiment is an embodiment in which lane formation determination is performed using simulation data 174, similar to the first embodiment. In the first embodiment, the lane formation determination device 10 performs the simulation, but in the second embodiment, a separately provided simulation server 20 performs the simulation. FIG. 12 is a hardware configuration diagram showing an example of implementing the lane formation determination device 10 in the second embodiment. The lane formation determination device 10 can be implemented by a computer, similar to the first embodiment, and includes a processing device 101, an input device 102, a display device 103, a communication device 104, a memory 105, and a secondary storage device 106, which are connected to each other via a communication path.

[0073] The lane formation determination device 10 is connected to a simulation server 20 that executes a simulation via a network 40. Therefore, the programs and information stored in the secondary storage device 106 are different from those in the first embodiment. That is, in the present embodiment, the secondary storage device 106 stores a lane formation determination program 110, layout data 171, person position information 172, and attribute information 173. These are the same as those in the first embodiment. Furthermore, at least a portion of the layout data 171, person position information 172, and attribute information 173 may be stored in the simulation server 20.

[0074] Similarly to the lane formation determination device 10, the simulation server 20 can also be realized by a computer. Therefore, the simulation server 20 has a processing device 201, a communication device 202, a memory 203, and a secondary storage device 204, which are connected to each other via a communication path. The secondary storage device 204 stores the simulation program 120 and simulation data 174. The processing device 201 executes a simulation of human movement in accordance with the simulation program 120, and generates the simulation data 174. In other words, the simulation server 20 corresponds to the simulation unit 16 in FIG. 2. Furthermore, the correspondence between these modules and programs and the units in FIG. 2 in this embodiment is the same as in the first embodiment.

[0075] Then, the communication device 202 transmits the simulation data 174 to the lane formation determination device 10 via the network 40. As a result, the lane formation determination device 10 can determine lane formation using the received simulation data 174. This concludes the description of the second embodiment. [Example]

[0076] In the first and second embodiments, lane formation determination is performed using simulation data 174, but in the third embodiment, actual measurement data 175 is used. FIG. 13 is a hardware configuration diagram showing an example of implementing the lane formation determination device 10 in the third embodiment. In FIG. 13, the lane formation determination device 10 is connected to a group of cameras 30 via a network 40. The lane formation determination device 10 is also connected to a group of terminal devices 50.

[0077] Furthermore, the lane formation determination device 10 can be realized by a computer, and includes a processing device 101, an interface device 107, a communication device 104, a memory 105, and a sub-storage device 106, which are connected to each other via a communication path. Here, the processing device 101, the communication device 104, the memory 105, and the sub-storage device 106 are the same as those in the second embodiment. However, compared to the second embodiment, the sub-storage device 106 further stores actual measurement data 175. This actual measurement data 175 is actual measurement data (images) of the route 1 captured by the camera group 30, and is acquired from the camera group 30. Then, lane formation determination is performed based on the actual measurement data 175.

[0078] The terminal device group 50 is also connected to the lane formation determination device 10 via the interface device 107, and displays the results to the user and accepts instructions and operations for determination. Therefore, the interface device 107 corresponds to the input / output unit 11 in FIG. 2. The terminal device group 50 displays the output contents shown in FIG. 10. The correspondence between these modules and programs and the units in FIG. 2 in this embodiment is the same as in the first embodiment.

[0079] Furthermore, the terminal device group 50 can be realized by a computer such as a PC, a smartphone, or a tablet. Although the description of each embodiment has been completed above, the present invention is not limited to these, and can be applied to various modifications and applications. For example, the scope of application of the present invention is not limited to the movement of people, but can also be applied to the movement of various mobile objects, such as their accessories and belongings. These mobile objects include mobility that transports people, such as robots and automobiles. It does not matter whether the movement of these mobile objects is autonomous or not.

[0080] Furthermore, even if there are obstacles or walls in the judgment area, it is possible to determine the formation of a detour lane by selecting neighboring cells in multiple directions (two or more directions) as search candidate cells or by using destination information and entrance information related to movement, as shown in Figures 3B and 3C. Furthermore, the present invention is not limited to two-dimensional people flow, but can also be applied to three-dimensional people flow.

[0081] Furthermore, the start cell group 1711 and the goal cell group 1712 may be interchanged to perform lane formation determination in both directions. [Explanation of symbols]

[0082] 1...route, 10...lane formation determination device, 11...input / output unit, 12...processing unit, 13...occupancy rate calculation unit, 14...attribute determination unit, 15...lane formation determination unit, 16...simulation unit, 17...storage unit, 171...layout data, 172...person position information, 173...attribute information, 174...simulation data, 175...actual measurement data, 1711...start cell group, 1711-1...start cell, 1712...goal cell group, 1713...determination area

Claims

1. an occupancy rate calculation unit that sets a lane formation determination area and calculates an occupancy rate indicating the degree of occupancy of each of a plurality of movement-related attributes in each cell that constitutes the determination area using position information of moving objects including people in the determination area; an attribute determination unit that identifies a cell attribute that represents each cell from the plurality of attributes according to the occupancy rate; A lane formation determination device having a lane formation determination unit that determines whether a lane is formed in the determination area based on the continuity of occupied cells having the same identified cell attribute.

2. 2. The lane formation determination device according to claim 1, The lane formation determination unit determines the continuity of the occupied cells based on whether or not cells with the same attribute are adjacent to each other in a row from a start cell to a goal cell in the determination area.

3. 2. The lane formation determination device according to claim 1, The lane formation determination unit is a lane formation determination device that determines that a lane has been formed within the determination area when a plurality of lanes with mutually different attributes are formed.

4. 4. The lane formation determination device according to claim 3, The attribute is a moving direction of the moving object.

5. 4. The lane formation determination device according to claim 3, The attribute is destination information indicating the direction of the destination of the moving object.

6. 2. The lane formation determination device according to claim 1, The occupancy rate calculation unit calculates the occupancy rate during a period in which the total proportion of different attributes within the determination area is equal to or greater than a certain value.

7. 2. The lane formation determination device according to claim 1, The lane formation determination unit is a lane formation determination device that, when it is determined that a lane is not formed in the determination area, resets a determination area that is included in the determination area.

8. In a lane formation determination method using a lane formation determination device, an occupancy rate calculation unit sets a lane formation determination area, and calculates an occupancy rate indicating the degree of occupancy of each of a plurality of movement-related attributes in each cell constituting the determination area using position information of moving objects including people in the determination area; an attribute determination unit identifying a cell attribute representing each cell from the plurality of attributes according to the occupancy rate; A lane formation determination method in which a lane formation determination unit determines whether a lane is formed in the determination area based on the continuity of occupied cells having the same identified cell attribute.

9. 9. The lane formation determination method according to claim 8, The lane formation determination method, wherein the lane formation determination unit determines the continuity of the occupied cells based on whether or not cells with the same attribute are adjacent to each other in a continuous manner from the start cell to the goal cell in the determination area.

10. 9. The lane formation determination method according to claim 8, A lane formation determination method that determines that a lane has been formed within the determination area when a plurality of lanes with mutually different attributes are formed by the lane formation determination unit.

11. 11. The lane formation determination method according to claim 10, A lane formation determination method in which the attribute is the moving direction of the moving object.

12. 11. The lane formation determination method according to claim 10, The lane formation determination method, wherein the attribute is destination information indicating the direction of the destination of the moving object.

13. 9. The lane formation determination method according to claim 8, The lane formation determination method, wherein the occupancy rate calculation unit calculates an occupancy rate within a period in which the total proportion of different attributes within the determination area is equal to or greater than a certain value.

14. 9. The lane formation determination method according to claim 8, A lane formation determination method for resetting a determination area that is included in the determination area when the lane formation determination unit determines that a lane is not formed in the determination area.

Images