JP2855157B2 - Crowd walking simulation system - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a crowd walking simulation system that sets a walking region and simulates a crowd flowing in the walking region.

[Conventional technology]

In the management and operation of event facilities, general leisure facilities, various events, and the like, it is necessary to analyze the number of people, their trends, and the trends of the crowd, and to provide appropriate guidance and guidance based on the analysis. In particular, in complex leisure facilities and mall-like commercial facilities, an unspecified number of people gather for various purposes. Therefore, it is desired to provide a variety of information according to each purpose for safety management and business management of a comprehensive facility group. To that end, it is essential to analyze trends after understanding the flow of people.

For example, in the case where the crowd is to be guided due to the stagnation of the crowd, the guide is conventionally guided by a placard, a megaphone, a rope, or the like, or by a guide broadcast by mobilizing human hands. As a conventional monitoring system for grasping the entire situation in this case, a television monitor system is often adopted.

[Problems to be solved by the invention]

However, a system that grasps the number of people, the trend of a crowd, and the like, analyzes the information in real time, and automatically performs guidance and the like as needed based on the analysis result has not been put into practical use.

The TV monitor system can relatively easily monitor the movements of the crowd, etc., but the observer always sticks in front of the monitor screen to monitor, and grasps the movement of the crowd etc. based on the intuitive judgment of the observer. Therefore, there is a problem that maintenance and operation require time and cost, as well as manpower, and it is not possible to respond quickly in real time. In particular, when the monitoring range is widened, there is a problem that the discovery of the occurrence of a serious emergency in management is overlooked or delayed at an early stage.

In addition, at facilities where visitors are to be managed, the entrance gate uses a method of counting people by means of a rotating type or a photoelectric tube type, but after entry, it is necessary to know how many people are on the premises, etc. Can not. Therefore, there is a method of counting the number of people using ultrasonic waves from the upper part of the detection area in order to grasp each area. However, this method has a problem that an error due to a change in weather is large.

In addition, there is a method using still image processing. However, this method requires a lot of labor and time because extraction of necessary information is performed manually for each image frame. is there.

The use of placards, megaphones, ropes, etc., or the use of guide broadcasting requires human resources for this purpose, especially in the case of severe mixing, which may cause dissatisfaction from the crowd because the guide may adversely affect traffic. Sometimes it becomes.
Also, it may be difficult to convey accurate instructions to the whole even in the case of emergency guidance. Furthermore, the method using the guide broadcast has a problem that it is difficult to be transmitted when the surroundings become noisy.

The present invention solves the above-mentioned problems, and an object of the present invention is to provide a crowd walking simulation system capable of grasping a trend of a crowd in real time and performing safety management and management and operation of a space where an unspecified number of people gather. That is. It is another object of the present invention to analyze the behavior of a crowd and perform appropriate processing according to the situation.

[Means for solving the problem]

Therefore, the present invention is a crowd walking simulation system that sets a walking area and simulates the flow of the crowd in the walking area, and includes pedestrians, obstacles, and pedestrians at each of destinations. It is characterized in that a distance-dependent repulsive force acts on an obstacle and a constant suction force irrespective of the distance acts between a pedestrian and a target point. The pedestrian and obstacles are given a positive charge and the destination is given a negative charge.The target walking area is represented by a magnetic field and simulated so that the pedestrian moves toward the destination based on Coulomb's law and equation of motion. It is characterized by doing.

[Action]

In the crowd walking simulation system of the present invention, pedestrians, obstacles, and pedestrians in each of destination points,
A distance-dependent repulsion force acts between the pedestrian and the obstacle, and a constant suction force regardless of the distance acts between the pedestrian and the target point. The flow of the crowd can be simulated with high accuracy by the equation of motion by the suction force.

〔Example〕

 Hereinafter, embodiments will be described with reference to the drawings.

FIG. 1 is a diagram for explaining one embodiment of a crowd walking simulation system according to the present invention, FIG. 2 is a diagram for explaining setting conditions of each element, and FIG. 3 is a diagram illustrating an example of a simulation process. FIG.

First, a basic concept of a model applied in the system of the present invention, a walking space (walking area), a walking principle, and a pedestrian will be described.

The basic idea of the model is to give a positive charge to each pedestrian and walls, pillars, handrails, etc. as shown in FIG. 2 (a) and a negative charge to the destination as shown in FIG. 2 (b). give. As a result, each pedestrian receives a repulsive force from other pedestrians and obstacles, and a suction force from the destination. The target pedestrian area is represented by a magnetic field, and the pedestrian moves toward the destination while avoiding other pedestrians and obstacles based on Coulomb's law and the equation of motion.

Regarding the representation of walking space, in order to faithfully reproduce the flow of people in areas with certain regularity in walking, such as around obstacles such as walls and pillars, near ticket gates, facility entrances, etc. A walking space is constructed using various magnetic charges. For example, the pedestrians are allocated by setting the magnetic dipole shown in FIG. 4D at the end of the partition wall near the entrance gate as shown in FIG. Without such magnetic dipoles, some pedestrians would stop completely within the edge circle. In addition, by setting the magnetic double layer shown in FIG. 7 (f) at the entrance to the room as shown in FIG. 7 (g), the pedestrian located in the vicinity of the end line circle can be smoothly introduced into the interior. Aspirate (induce).

The walking area is represented by a portion surrounded by connecting arbitrary vectors representing a wall in a clockwise direction as shown in FIG. In addition, sequential numbers are assigned so that the contact numbers of the connection points (vertexes) of the wall vectors are always clockwise when viewed from the walking area. This allows
It is possible to determine whether or not the pedestrian is in a normal position (located in the walking area).

In addition, the connection points protruding toward the walking area among the connection points of the wall vectors, for example, in the example shown in FIG.
13, 14, 18, 20, 21, 23 to 26 are recognized as “corners”.

The position in the walking area is defined by two-dimensional coordinates (XY).

 The walking principle is handled as follows.

Each pedestrian has a territory with a radius of 3 m, as shown in FIG. 3 (i), and receives a repulsive force only when another pedestrian or a disabled person enters the traveling direction 180 °. Further, a constant suction force is received from the destination regardless of the distance. The suction force does not follow Coulomb's law. If the pedestrian receives a plurality of forces, it is expressed as a result of summing (vector sum) the respective forces as shown in FIG.

 Pedestrian expressions are handled as follows.

Pedestrian groups of multiple groups with different walking routes are set based on the conditions given, and the walking speed, occurrence / final destination, occurrence density, and group size (20 to 110 people) are set for each group.
And other parameters are defined. That is, each pedestrian belongs to any group, and the pedestrian data is fixed data as a group, and is given for each group. The group data includes the walking speed, the occurrence position, the occurrence density, the number of people in the group, and the walking route. Data (speed, occurrence position) of each pedestrian in an arbitrary group is automatically given by random numbers based on the data of the group.

The input format of the pedestrian data of each simulation is, for example, as follows.

1st line: whether pedestrian data is group input or individual input (= 0 indicates group input) 2nd line: number of groups 3rd line: destination, search format, occurrence position vector walking speed, occurrence density, occurrence time 4th line: Number of path display contact points 5th line: Path display contact number (continuous 3 digits) Hereinafter, data of one group is input every 4 lines.

As for the walking route, a plurality of points as a migration route on the way are designated in addition to the occurrence / final destination. These are given as a sequence of wall contact numbers, known as corners. However, this is because the walking characteristics of a human being "attempt to travel the shortest distance to the destination while receiving interference from obstacles and other pedestrians" are taken into consideration.

The walking speed is given the average speed of a human under good walking conditions, and has some variation by random numbers based on the average speed.

The generation density is given by the setting conditions of each simulation pattern.

The pedestrian recognizes the designated corner as the destination after the occurrence within the walking area, and receives the suction force from the point while proceeding with interference (repulsive force) from obstacles and other pedestrians. After passing through, the same cycle is repeated every unit time until reaching the destination point at the next designated corner.

Here, as a special element of the simulation model related to the pedestrian space, for example, “Pedestrian occurrence / final destination”
There are a "gate to enter the Aqua Museum" and a "viewing / watching point". At these points, pedestrians act based on logic different from the above-described walking principle.

"Pedestrian occurrence / final destination point" means that a pedestrian is generated or arrives not on a point but on a line connecting the vertices of the wall. -Reach) is determined.

At the “Aqua Museum entrance gate”, after each pedestrian arrives, the pedestrian stays at that point until receiving a predetermined service (corresponding to a waiting time).
If a queue has already been formed at the time of arrival, the queue will be at the end of the shortest row.

The “viewing / viewing point” is given by setting a plurality of windows at equal intervals on a line connecting the vertices of the wall. When all the windows are full, a queue is formed similarly. However, in the case of the "entrance gate", the service starts moving forward to the next destination after the service ends, but in this case, it goes around the end of the queue and goes to the next destination, so once, A temporary destination is set near the end of the queue, and after passing through the destination, the user goes to the original next destination.

Since these three types of special elements have a linear shape with respect to the walking area, they cannot be represented by wall contact numbers. Therefore, a sequential number is newly assigned to a line connecting adjacent contacts, and parameters such as a service time, the number of gates, and the number of windows are set for the corresponding portion.

Next, the procedure (flowchart) of the simulation calculation will be described in detail with reference to FIG.

(S1) A group of pedestrians is actually generated, and various environmental data are set before simulating their actions. First, facility plane layout data for setting a walking area and data on a pedestrian group for each group described above are input.

Pedestrian data is divided into fixed data as a group to which pedestrians belong and pedestrian files containing variable data possessed by each person during walking behavior. Here, the former group data (18 patterns) shown below Is input.

Group size, occurrence / final destination, walking speed,
Occurrence density, occurrence time, route display contact number (migration route).

(S2) It is recognized whether or not the connection points (vertexes) of all the walls are corners (projecting with respect to the walking area).
In the determination method, when sin θ> 0 with respect to the angle θ formed by the two vectors forming the connection point of the wall, the connection point is recognized as a corner. Therefore, for example, the third
In the example shown in FIG. 5A, each of θ _{1 to} θ ₅ is
When the value of siθ is obtained, a and c satisfying sinθ ₁ > 0 and sinθ ₃ > 0 are recognized as corners.

(S3) Recognize the adjacency between vertices recognized as corners in the above processing. This is referred to when determining a tentative target point when the pedestrian enters a search walking state described later. The adjacency relationship is “connected” when one vertex looks at another vertex and satisfies the following conditions.
Is determined.

What can be seen Only one side of the opponent's vertex can be seen This adjacency relationship is represented by the following matrix in the walking space shown in FIG. 3B, for example.

Here, it is “1” when connected, and “0” when not connected. Since this is a directional flag, the adjacency matrix is not necessarily a target matrix.

(S4) A pedestrian is generated in the walking area, and position coordinates are calculated for all walking steps per unit time based on the above-described walking principle. At this time, the calculation of the magnetic force applied to the pedestrian is performed for each pedestrian. Therefore, points other than the pedestrian's destination point that is the target of calculation do not have a negative charge so that pedestrians belonging to one group do not receive suction force from the destination point for pedestrians in another group. To do.

(S5) The temporary destination of each pedestrian is determined, and the walking route is checked.

The processing at the time of the corner passage in the check of the walking route is performed as follows.

The pedestrian receives the suction force from the designated corner or a point on the line segment, and reaches the final destination by sequentially passing through them. However, if the service is not actually received, that is, as shown in FIG. 3 (c), if a corner (mostly applicable) designated to walk the shortest route along the migration route has a negative charge, The pedestrian will exhibit an unnatural movement such as "not hitting a wall, but hitting a wall (corner) and then heading to the next corner". Therefore, in practice, a natural flow is realized by setting a tentative destination point slightly inside the designated walking area of the corner as shown in FIG. 3 (d).

Regarding the recognition of the passage of the destination point, as shown in FIG. 3 (e), "the pedestrian passed when the walls on both sides constituting the corner were visible (the shaded area in the figure)."
However, in actual calculation, the following method is used.

An angle θ formed by a vector A connecting the coordinates where the pedestrian is located and the coordinates of the destination and a wall vector B from the destination is determined. Then, as shown in FIG.
In the case of 0, it has not passed, and as shown in FIG.
If θ ≦ 0, it is determined that the vehicle has passed.

(S6) For example, a queuing occurs at the “amusement entrance gate” or the “viewing / viewing point”. When the pedestrians are in a queue, they have a different recognition from the normal walking state, and do not follow the above-described walking principle. In this case, as shown in FIG. 3H, a virtual point matrix is set at intervals of 50 cm in front of each window provided on the line segment, and pedestrians are sequentially arranged on this. Each pedestrian secures a position (virtual point) where the waiting time is shortest when the window is seen, that is, the length of the queue is minimum. At that time, the corresponding virtual point (●) becomes a temporary destination for the pedestrian, and is sucked by having a negative point charge.

(S7)-(S8) Speed from magnetic force received for each pedestrian,
Calculate direction.

(S9) Each pedestrian is displayed on the screen together with the walking area.

(S10) Store the data in the pedestrian file.

In this data storage process, the state change for each unit time is written in the pedestrian file containing the variable data possessed by each of the above-mentioned walking actions. The pedestrian file is composed of the following data, for example.

Position coordinates, walking speed direction, size and upper limit,
Final destination point, next destination point, walking route (route display contact number), already passed point, walking style (route dependent walking or
Search walking), walking state (not generated or occurred, normal walking, queue (close to queue, in queue, leaving queue,
In addition, some pedestrians lose track of the next destination because they received interference (repulsion) from other pedestrians during the simulation. At that time, the pedestrian enters a state of searching and walking, and sets an optimal corner as a temporary destination based on the current position coordinates and the position coordinates of the destination.

The criterion for selecting the optimum corner is as follows.

Pedestrians remember the corners they have passed once, and in principle never go there again. This is to prevent a pedestrian from falling into an infinite loop. Therefore, an optimum corner is selected from corners that have not passed yet.

With reference to the adjacent relation matrix between corners of the destination point that has been lost, a corner having an adjacent relation is selected from the matrix.

Next, of the two walls that make up the corner, select the one that can only be seen on one side. If all corners in front of the pedestrian show walls on both sides,
From there, there is a dead end.

Pedestrians generally choose the corner in front of their direction of travel, but if the front is a dead end, or if the front corner has just passed, the rear corner will be used. Make it a selection target.

When there are a plurality of corners satisfying these conditions, a corner closest to the lost destination is selected as a temporary destination.

Each time the pedestrian passes through a corner selected as a tentative destination while repeating such a search and walk, the pedestrian determines whether or not a lost destination can be seen from that point. If it can be seen, it recognizes that it has returned to the normal walking state. With this logic, it is possible to escape from search walking on the shortest route.

For example, when there is a walking space as shown in FIG. 3 (i), an area where a non-passing pedestrian can visually recognize the destination is a shaded area. At this time, a pedestrian who has entered the area (a) due to interference from another pedestrian or the like loses sight of the next destination. This determination is made based on the fact that the vector from the pedestrian to the destination crosses an arbitrary wall vector. At this time, the pedestrian enters a state of searching and walking, and selects an optimum temporary destination from the corners Q, R, and K.

In other words, the corner K that satisfies all of the following conditions (not relevant in this case) is the temporary destination. It should be noted that the determination of the corner in front of the pedestrian is made by calculating the inner product of the vectors. For example, the corners R and K are PNPNPK> O and PN ・ PR> 0, and both are positive, so they are considered to be ahead. For the corner Q, PN · PQ <0, and it is determined that the vehicle is not ahead.

Next, an example of a simulation using Coulomb's law and equation of motion under the conditions set as described above will be described.

Force F ₂ to receive from the force F ₁ and the other pedestrian received from pillar than the Coulomb's law is expressed as follows. Incidentally, the suction force F ₀ from destination point is constant without following Coulomb's law.

r ₁ : Distance from point P to column Q ₁ : Charge of column r ₂ : distance from point P to pedestrian Q ₂ : electric charge of pedestrian ε ₀ = 8.855 × 10 ^{-12 In} general, when a unit vector is used for force F, F = F _x i + F _y j (3) i: x unit vector j: y direction unit vector x ₀ , y ₀ ) Since the relationship is the same for F _y , F can be expressed as follows.

If the above equations (1) and (2) are expressed according to the equation (4), F ₁ and F ₂ are Becomes However, since the origin of the coordinates is at a point where the pillar and other pedestrians are located, it is necessary to correct the origin. The coordinates of the pillar (x _1, y _1), the other a pedestrian coordinate (x _2,
y ₂ ), the expressions (5) and (6) are Force F ₃ is F _1, F ₂ and _{F F 3 = F 1 + F} 2 + F 0 for representable vector sum of ₀ ... (9) and at the point P, x-direction, is decomposed by the y-direction component F ₃ = to _{(F x1i + F y1j) +} (F x2i + F y2j) + (F x0i + F y0j) = (F x1 + F x2 + F x0) i + (F y1 + F y2 + F y0) j ...... (10) (10) formula Substituting equations (7) and (8) gives Since the force applied to point P is not always three, (1
1) The general solution of the equation is as follows.

l: An obstacle in the pedestrian's territory m: Another pedestrian in the pedestrian's territory The pedestrian proceeds in accordance with the equation of motion by the force F expressed by the equation (12).

m: Pedestrian mass a: Pedestrian acceleration L: Pedestrian displacement (absolute value) V: Pedestrian speed l: Initial value Similarly, by equation (13) Substituting equation (15) into equation (14) Here, assuming that the unit time is 1 (t = 1) and the initial value is 0 (l = 0) in order to obtain the absolute displacement, the equation (16) becomes: If the initial value of V is V ₀ , V is expressed as a function of t,
To become V (t) = V (t -1) + V 0. For V, an upper limit is set, and a check is made so that the walking speed does not become extremely high.

Decomposing the absolute displacement amount L into y components in the x direction and direction and substituting equation (12) gives Therefore, when the position coordinates of the pedestrian at time t are (x ₀ , y ₀ ), the position (x ₁ , y ₁ ) at (t + ₁ ) is The applicant has proposed a moving body identification analysis management system that separately identifies pedestrians by image processing, analyzes the trend, and processes the pedestrians, but the present invention uses this system in combination with a comprehensive leisure land and trade fair. It can contribute to the management of large-scale exhibition halls and the like. Hereinafter, the moving object identification analysis management system will be described.

FIG. 4 is a diagram showing an embodiment of a moving object identification analysis management system, FIG. 5 is a diagram showing an example of an analysis model, and FIG. 6 is a diagram for explaining a processing procedure.

In FIG. 4, image memories 2 and 3 store video data (image data). The image data is obtained by photographing a monitoring area from above using an imaging device such as a CCD camera or a video camera. Things. The switching unit 1 stores the input image data in the image memories 2 and 3 every predetermined time.
In such a manner as to be stored alternately. The moving object extraction unit 4 extracts a moving object by taking the difference between the images stored in the image memories 2 and 3, and the image memories 5 and 6 store image data obtained by extracting the dynamics. The position extracting unit 7 extracts the position of the dynamic at each time from the image data of the dynamic in the image memories 5 and 6, and the analyzing unit 8 calculates the position of the moving object in the monitoring area from the temporal change of the extracted position of the dynamic. The movement is analyzed, or the number of moving objects is counted, and a trend is analyzed based on a set of moving objects.

 Next, the processing procedure will be described with reference to the model shown in FIG.

In the model shown in FIG. 5, a monitoring area 11 is photographed from above by a camera 13, and image data including a moving object 12 such as a human being is taken into an image processing device 14. The image processing device 14 extracts the image data of the moving object 12 by calculating the difference between the preceding and succeeding image data at regular intervals, and further performs position extraction processing of the moving object 12. Then, the extracted position data is transferred to the host computer 15, and data analysis is performed. That is, in the model shown in FIG. 5, up to the position extraction unit 7 shown in FIG.
The rest is constituted by the host computer 15.

The process performed by the moving body extracting unit 4 and the position extracting unit 7 will be described. First, the switching unit 1
Images f1, f2, and f3 are captured in the order of FIGS. The switching unit 1 stores the first image f1 in the image memory 2
And the image data is switched so that the next image f2 is stored in the image memory 3. Therefore, the dynamics extraction unit 4 reads the image data stored in the image memories 2 and 3 to obtain the difference between the images f1 and f2, and stores the image m12 in the image memory 5. The image m12 obtained as the difference between the images f1 and f2 has only the changed part as shown in FIG. 6D, that is, the information of the moving body (person) 12 whose position is not the same in the images f1 and f2. Remains. Background portions such as roads and surrounding stationary objects overlapping in the images f1 and f2 are images f1 and f2.
Since there is no change in the same position at f2, the difference is eliminated by taking the difference. In this case, even in the case of the information of the moved person 12, a part of the overlapping portion between the images f1 and f2 may be partially removed, but this can be compensated for by processing such as filling in and expanding.

Further, when the third image f3 is captured, the switching unit 1
Then, the image data is switched again so that the image f3 is stored in the image memory 2. Then, similarly, the dynamics extraction unit 4
Then, the image data stored in the image memories 2 and 3 are read, and the difference between the images f2 and f3 is obtained, thereby obtaining an image m23 shown in FIG.
When the difference images m12 and m23 are stored in the image memories 5 and 6, respectively, the position extraction unit 7 obtains a logical product of the images m12 and m23 to obtain a position extraction image a123 as shown in FIG. ,
The position where the moving body 12 was located at the time of the image f2 is extracted.

In the analysis unit 8, as shown in FIGS.
When the time-sequential position extraction images a012, a123, and a234 are obtained as shown in (1), the speed and trend are analyzed from the direction and distance of movement by comparing the positions of the centers of gravity.

The above processing is continuously performed at regular time intervals, and the captured images f1, f2,..., M12, m23,.
By analyzing a234,..., the individual and overall trends of the moving body (person) are analyzed. In analyzing this trend, for example, the number of moving objects and a change in the number of moving objects are obtained by counting the number of moving objects, and a change in the overall moving direction and speed is obtained.

As described above, images are captured with a time difference from the same angle, and changes are extracted by taking the difference between those images, so that only the image information of the moving object is extracted regardless of the environment where the moving object is placed Further, by taking a logical product between the difference images, the position of the moving object can be extracted at each time point. Therefore, by counting this, the number of moving objects is obtained, and for example, the waiting time can be calculated or the occurrence of an emergency can be detected from the change in the number, the direction of the overall movement, the change in the speed, and the like.

7 and 8 are diagrams for explaining an example in which a camera is installed at an entrance and a trend is analyzed.

As shown in FIG. 7, when the vicinity of the entrance or the exit of the facility is set as a monitoring area, the monitoring area 22 is photographed by the camera 21 from above, and the above processing is performed on the image data at regular time intervals.
For example, at the entrance and the exit, the number of visitors and the number of exits per unit time and the flow (speed) thereof can be obtained. Furthermore, when the number of visitors at the entrance is added and the number of exits at the exit is subtracted, the number of visitors at each time, the increase / decrease, the staying status, and the like can be obtained. In addition, if the same analysis is performed on the entrance and the area on the way to the entrance, the number of people lined up at the entrance can be obtained, and the waiting time can be calculated from the number of people and the flow (speed) at the entrance / exit of the venue.

At the entrance, the moving direction of the moving body is determined as shown in FIG. Therefore, by monitoring the moving direction and the moving speed, it is possible to detect the occurrence of an abnormal state in the venue. As a pattern of the abnormal state, for example, a case where the moving speed suddenly increases or stops suddenly, or a case where the flow is reversed at the entrance.
Therefore, if an abnormality is detected by the system from such a change in the flow or the pattern of the direction, the details of the abnormality cannot be grasped. Therefore, as shown in FIG. It is possible to quickly take countermeasures such as checking the content of the abnormality. In addition, it is also possible to set a no-go area or the vicinity of an emergency exit as a monitoring area, detect the approach of a moving object, and determine an abnormality. The criteria for the determination may be changed depending on the time zone. Since these are judgments of abnormalities due to the movement of the moving body, the actual situation is confirmed by a staff member confirming the site or a monitor screen.

FIG. 9 is a diagram showing a configuration example of a visitor monitoring system in a wide area facility.

In FIG. 9, a video switcher 32 is for switching video signals of the surveillance cameras 31-1 to 31-18, and a time generator 36 is a video switcher 32.
, A calculation of the personal computer 34, and a timing signal such as sampling for each area in the logger 35. The image processing device 33 and the personal computer 34 are the same as the image processing device 14 and the host computer 15 described above with reference to FIG. 5, and the logger 35 uses the number of persons obtained by the personal computer 34 as sampling data for each area. It accumulates and memorizes. The calculating unit 37 calculates the number of persons in a predetermined section by adding the number of persons on the incoming side and subtracting the number of persons on the outgoing side from the sampling data for each area accumulated in the logger 35, for example, as described above. It is. The data storage unit 38
The calculation data of the number of persons in the section is stored, and the trend analysis unit 39 analyzes the data in each section stored in the data storage unit 38 to analyze and predict the staying state.
That is, when the number of persons in a certain section exceeds a certain value, it is determined that the staying state is present, and when the number of persons in the section exceeds a certain ratio, the staying state is predicted. The display device 40 outputs the staying status analyzed by the trend analysis unit 39.

In addition, the trend analysis unit 39 predicts a stagnation state for each area, thereby setting a flow pattern for eliminating the stagnation, and outputting the pattern as guidance information to the display device 40 or other display boards. You may comprise. The setting of the pattern in this case may be configured such that a selection pattern responding to the staying state of each area is prepared in advance, and a selection pattern is selected from the selection pattern under certain conditions.

The crowd walking simulation system of the present invention can predict the situation after a certain period of time in real time by using the information of the pedestrian and the speed obtained by the above system, and provide a sign at an optimal place. Can be.

It should be noted that the present invention is not limited to the above embodiment, and various modifications are possible. For example, in the above-described embodiment, the repulsion force and the attraction force are set by giving a positive charge and a negative charge. However, in a broad sense, the charge in the embodiment is transferred by a moving force factor, a magnetic field or a magnetic dipole. Needless to say, by grasping the magnetic double layer or the like as a driving force factor, these factors may be replaced with other factors and the repulsive force and the attractive force may be set.

〔The invention's effect〕

As is clear from the above description, according to the present invention, a pedestrian is simulated using a repulsive force and a suction force, so that crowd behavior can be analyzed and predicted easily and with high accuracy. Furthermore, a flexible sign device for crowd behavior management can be realized by combining with a separately proposed moving body identification analysis management system. Therefore, it is possible to efficiently perform crowd safety management and business management in various large-scale venues and facilities.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining an embodiment of a crowd walking simulation system according to the present invention, FIG. 2 is a diagram for explaining setting conditions of each element, and FIG. 3 is a diagram for explaining an example of a simulation process. FIG. 4 is a diagram showing an embodiment of a moving object identification analysis management system, FIG. 5 is a diagram showing an example of an analysis model, FIG. 6 is a diagram for explaining a processing procedure, FIG. 8 is a diagram for explaining an example of installing a camera at an entrance and analyzing a trend, and FIG. 9 is a diagram showing a configuration example of a visitor monitoring system in a wide area facility. 1 switching unit, 2, 3, 5, and 6 image memory, 4
Moving object extraction unit, 5: position extraction unit, 8: analysis unit.

Continued on the front page (72) Inventor Hidehiko Nakagawa 2- 16-1, Kyobashi, Chuo-ku, Tokyo Shimizu Construction Co., Ltd. (72) Inventor Toyo Daimon 2-6-1 Kyobashi, Chuo-ku, Tokyo Shimizu Construction Co., Ltd. (72) Inventor Toyokatsu Sato 2-16-1, Kyobashi, Chuo-ku, Tokyo Shimizu Construction Co., Ltd. (72) Inventor Keigo Takeuchi 2-16-1, Kyobashi, Chuo-ku, Tokyo Shimizu Construction Co., Ltd. (56) Reference Document JP-A-49-89088 (JP, A) JP-A-60-205664 (JP, A) JP-A-59-142842 (JP, U) Keiichiro Ando (Two outsiders) "Predicting the flow of people" Passenger Flow Simulation System ”RRR Japan Railway Technical Research Institute 1988.8.12 Vol. 45 No. 8 p. 8-13 Hitoshi Watanabe, “Crowd and Evacuation,” Architectural Disaster Prevention Japan Architectural Disaster Prevention Association of Japan 1987. 4.6 No. 112 p. 2-8 (58) Fields surveyed (Int. Cl. ⁶ , DB name) G08G 1/00 G06F 15/20 G09B 9/00 JICST file (JOIS)

Claims (8)

(57) [Claims] 1. A crowd walking simulation system for setting a walking area and simulating the flow of a crowd in the walking area, wherein a pedestrian, an obstacle, and a destination point each other, A crowd walking simulation system characterized in that a distance-dependent repulsive force acts on an obstacle and a constant suction force acts on a pedestrian and a target point regardless of the distance. . 2. A pedestrian and an obstacle are given a positive charge and a destination is given a negative charge, and a target walking area is represented by a magnetic field so that the pedestrian proceeds toward the destination based on Coulomb's law and the equation of motion. The crowd walking simulation system according to claim 1, wherein the simulation is performed. 3. The apparatus according to claim 1, wherein the pedestrian is set to receive a repulsive force only when another pedestrian or an obstacle enters the traveling direction 180 ° with a constant radius. Crowd walking simulation system. 4. An arbitrary vector is connected to represent a wall,
2. The crowd walking simulation system according to claim 1, wherein the connection point is recognized as a corner when a value of a sine regarding an angle formed by two vectors forming the connection point of the wall is positive. 5. The crowd walking simulation system according to claim 1, wherein a corner in the middle of the destination point is sequentially simulated as a temporary destination point up to a final destination point. 6. The crowd walking simulation system according to claim 5, wherein a temporary target point is set on a side of the corner where the walking area is located. 7. The crowd walking simulation system according to claim 5, wherein the passing of the temporary target point is recognized on condition that the walls on both sides of the temporary destination point are visible. 8. The crowd walking simulation system according to claim 7, wherein the determination of passage is made based on an angle between a vector connecting the coordinates of the pedestrian and the coordinates of the destination and a wall vector from the destination. .