Crowd evacuation simulation method, device and terminal equipment

Abstract

The application is suitable for the technical field of crowd simulation, and provides a crowd evacuation simulation method, device and terminal equipment, which comprises the following steps: constructing a scene model; constructing a pedestrian model and an attacker model based on the scene model, wherein the pedestrian model comprises a pedestrian motion rule, and the attacker model comprises an attacker motion rule; constructing an interaction model based on the pedestrian model and the attacker model, wherein the interaction model comprises multiple interaction modes, and the multiple interaction modes can be converted to each other; and performing crowd evacuation simulation based on the scene model, the pedestrian model, the attacker model and the interaction model. The application can more realistically simulate the crowd evacuation movement in a terrorist attack, and provides data support and scientific basis for the construction and management of public places, so as to reduce the casualties of the crowd in a terrorist attack.

Description

Technical Field

[0001] This application belongs to the field of crowd simulation technology, and in particular relates to a crowd evacuation simulation method, device and terminal equipment. Background Technology

[0002] Terrorist attacks refer to deliberate attacks perpetrated by extremists or organizations against, but not limited to, civilians. They are not only highly lethal, causing immense casualties and property damage, but also triggering social unrest and hindering economic development. Such attacks often occur in crowded public places where evacuation is difficult.

[0003] Due to the unique nature of terrorist attacks and their immense danger, crowds become unstable, and escape routes become even more complex when attacked by terrorists, making realistic simulated evacuation drills impossible.

[0004] Therefore, there is an urgent need for a simulated evacuation model for terrorist attacks, which can more realistically simulate the evacuation of people during terrorist attacks, providing data support and scientific basis for the construction and management of public places, so as to reduce casualties. Summary of the Invention

[0005] To overcome the problems existing in related technologies, this application provides a crowd evacuation simulation method, device and terminal equipment, which can more realistically simulate crowd evacuation movements during terrorist attacks, provide data support and scientific basis for the construction and management of public places, and reduce crowd casualties in terrorist attacks.

[0006] This application is achieved through the following technical solution:

[0007] In a first aspect, embodiments of this application provide a crowd evacuation simulation method, including:

[0008] Construct a scenario model; based on the scenario model, construct a pedestrian model and an attacker model. The pedestrian model includes pedestrian movement rules, and the attacker model includes attacker movement rules. Based on the pedestrian model and the attacker model, construct an interaction model. The interaction model includes multiple interaction modes, which can be converted into each other. Based on the scenario model, pedestrian model, attacker model, and interaction model, conduct crowd evacuation simulation.

[0009] In one possible implementation of the first aspect, multiple interaction modes include a search mode, a chase mode, and an attack mode. In the search mode, the attacker locates the nearest pedestrian based on attacker movement rules; the search mode ends once the attacker determines the nearest pedestrian's location. In the chase mode, the attacker chases the pedestrian based on attacker movement rules, while the pedestrian moves away from the attacker based on pedestrian movement rules. In the chase mode, the attacker can update the chase path based on the real-time changing positions of both the attacker and the pedestrian; the chase mode ends once the attacker catches up with the pedestrian. In the attack mode, after the attacker catches up with the pedestrian, both remain stationary until the pedestrian dies after a preset time; the attack mode ends upon the pedestrian's death. The attacker locates the pedestrian's position through the search mode, enters the chase mode, catches up with the pedestrian through the chase mode, and then enters the attack mode. After the attack mode ends, the attacker re-enters the search mode.

[0010] In one possible implementation of the first aspect, when the attacker is chasing a target pedestrian in the chase mode, the shortest path pedestrian position is updated in real time based on the real-time changing attacker position and the real-time changing pedestrian position, and the target pedestrian is changed based on the shortest path pedestrian position.

[0011] In one possible implementation of the first aspect, the scene model is a three-dimensional scene model. Constructing the scene model includes: setting the scene environment, which includes scene size, scene height, evacuation exit locations, evacuation exit sizes, obstacle models, number of characters, and initial positions of the characters. The obstacle model includes obstacle locations, obstacle shapes, and obstacle heights. The scene size is divided into multiple grids based on a preset scale, with each grid representing a cell.

[0012] In one possible implementation of the first aspect, pedestrian motion rules are constructed based on a scene model, including:

[0013] Based on the scene model, the pedestrian Moore neighborhood, static ground field, and dynamic ground field are determined. The pedestrian Moore neighborhood is the set of the pedestrian position and the 8 cells adjacent to the pedestrian position. The static ground field is determined by the shortest path calculated based on the evacuation exit position and obstacle model. The larger the static ground field value, the easier it is for pedestrians to evacuate. The dynamic ground field is determined by the shortest path calculated based on the attacker position and obstacle model. The larger the dynamic ground field value, the farther away from the attacker position, and the easier it is for pedestrians to evacuate.

[0014] Based on the pedestrian's Moore neighborhood, static ground field, and dynamic ground field, the transition probability of a pedestrian in the Moore neighborhood is determined. The transition probability expression is: P ij =N[exp(k s S ij +k d D ij (1-n) ijIn the formula, P ij S represents the transition probability of a pedestrian in cell (i, j), N represents the normalization factor, N∈[0,1], and S ij Let k represent the static ground field when a pedestrian is in cell (i, j). s k represents the sensitivity coefficient of the static ground field. s ∈[0,1],D ij Let k represent the dynamic ground field of a pedestrian in cell (i, j). d k represents the sensitivity coefficient of the dynamic ground field. d ∈[0,1],n ij Indicates a placeholder, n ij It is 0 or 1 when n ij =0 indicates that cell (i, j) is occupied, and the pedestrian cannot move to this cell. When n ij =1 indicates that cell (i, j) is not occupied and pedestrians can move to this cell.

[0015] The pedestrian's moving target position is obtained based on the transition probability. The cell with the maximum transition probability is the pedestrian's moving target position.

[0016] A solution is defined for conflicting destination locations. The solution is that if multiple pedestrians have the same destination location, the pedestrian with the highest probability of moving to that destination location moves there, while the other pedestrians move to the location with the highest probability of moving among all other locations besides the destination location.

[0017] In one possible implementation of the first aspect, attacker movement rules are constructed based on a scenario model, including:

[0018] The attacker's field of vision is determined based on the obstacle model and the attacker's field of vision model. The attacker's field of vision model is a fan-shaped area at a preset angle directly in front of the attacker when there are no obstacles obstructing the view.

[0019] Based on the attacker's field of vision, attacker's location, and pedestrian's location, the maximum attraction of the pedestrian to the attacker is determined. The expression for the maximum attraction is:

[0020]

[0021] In the formula, F pa The value represents the maximum attractiveness of a pedestrian to an attacker, k represents a constant, and x represents the maximum attractiveness of a pedestrian to an attacker. p and y p Let x be the position coordinates of pedestrian p. a and y a Let be the coordinates of attacker a, where pedestrian p is within attacker a's field of vision.

[0022] The attacker's movement rules are determined based on the maximum attraction of pedestrians to the attacker, and the attacker's movement trajectory follows the A-star algorithm.

[0023] In one possible implementation of the first aspect, a trigger is set in the attacker model. When a preset condition is met, the trigger switch is activated, and the attacker begins to execute the attacker's movement rules.

[0024] Secondly, embodiments of this application provide a crowd evacuation simulation device, including: a scene model building module, a person model building module, an interaction model building module, and a simulation module. The scene model building module is used to construct a scene model; the person model building module is used to construct a pedestrian model and an attacker model based on the scene model, the pedestrian model including pedestrian movement rules, and the attacker model including attacker movement rules; the interaction model building module is used to construct an interaction model based on the pedestrian model and the attacker model, the interaction model including multiple interaction modes that can be converted between each other; and the simulation module is used to perform crowd evacuation simulation based on the scene model, pedestrian model, attacker model, and interaction model.

[0025] Thirdly, embodiments of this application provide a terminal device, including a memory and a processor. The memory stores a computer program that can run on the processor. When the processor executes the computer program, it implements the crowd evacuation simulation method as described in any of the first aspects.

[0026] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the crowd evacuation simulation method as described in any of the first aspects.

[0027] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the crowd evacuation simulation method described in any one of the first aspects.

[0028] It is understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0029] The beneficial effects of the embodiments in this application compared with the prior art are:

[0030] This application embodiment constructs a scene model, a pedestrian model, an attacker model, and an interaction model. By setting pedestrian movement rules, attacker movement rules, and various interaction modes based on pedestrians and attackers, it simulates crowd evacuation. This application embodiment provides a novel crowd evacuation method that can more realistically simulate crowd evacuation in a three-dimensional terrorist attack scenario. It considers both the movement and interaction of pedestrians and attackers, as well as the impact of different obstacles on crowd movement. This can provide data support and scientific basis for the construction and management of public places, as well as for counter-terrorism organizations, to reduce casualties in terrorist attacks.

[0031] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a schematic flowchart of a crowd evacuation method provided in an embodiment of this application;

[0034] Figure 2 This is a schematic diagram of the process for constructing pedestrian movement rules according to an embodiment of this application;

[0035] Figure 3 This is a schematic diagram of the range of the mole neighborhood and the direction of pedestrian movement provided in an embodiment of this application;

[0036] Figure 4 This is a schematic diagram of the process for constructing attacker movement rules provided in an embodiment of this application;

[0037] Figure 5 These are three initial scene diagrams for Experiment 1 provided in one embodiment of this application;

[0038] Figure 6 This is a line graph showing the number of pedestrians versus the number of deaths in Experiment 1, provided in one embodiment of this application.

[0039] Figure 7 These are two initial scene diagrams for Experiment 2 provided in one embodiment of this application;

[0040] Figure 8 This is a line graph showing the number of pedestrians versus the number of deaths in Experiment 2, provided in one embodiment of this application.

[0041] Figure 9 These are two initial scene diagrams for Experiment 3 provided in one embodiment of this application;

[0042] Figure 10 This is a line graph showing the number of pedestrians versus the number of deaths in Experiment 3, provided in one embodiment of this application.

[0043] Figure 11 This is a schematic diagram of the structure of the crowd evacuation simulation device provided in the embodiments of this application;

[0044] Figure 12 This is a schematic diagram of the structure of the terminal device provided in the embodiments of this application. Detailed Implementation

[0045] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0046] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0047] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0048] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] The dangerous acts of terrorists cause immense public panic and have a very negative impact on social stability. According to statistics from the Global Terrorism Database (GTD), the number of deaths caused by terrorist attacks worldwide reached 20,300 in 2019 alone, and terrorist attacks have shown a severe trend of rapid spread globally in recent years. Due to the unique nature of terrorist attack scenarios, realistic simulations are impossible; therefore, computer simulations of emergency evacuation are an effective means of analyzing the number of casualties from terrorist attacks.

[0051] Existing simulations of crowd evacuation cannot accurately simulate crowd evacuation under terrorist attacks. On the one hand, due to the dangerous behavior of attackers, pedestrians will try their best to ensure their safety and stay away from the attackers, thus causing disorderly movement of the entire crowd; on the other hand, terrorist attacks generally occur in densely populated scenes, which are complex environments with many obstacles of varying shapes and heights, and these obstacles have a significant impact on the escape routes and time of the crowd during evacuation.

[0052] To address the aforementioned problems, this application provides a crowd evacuation simulation method. To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described below are merely illustrative of this application and are not intended to limit its scope.

[0053] Please see Figure 1 A flowchart illustrating a crowd evacuation method according to an embodiment of this application is provided. The crowd evacuation method according to an embodiment of this application may include the following steps:

[0054] In step 101, a scene model is constructed.

[0055] In some embodiments, a scene model is constructed based on cellular automata. This scene model expands the two-dimensional plane into a three-dimensional space based on the traditional ground CA model.

[0056] For example, in order to make the scene model more similar to the real environment, not only the scene size, scene height and evacuation exit location were set, but the size of the evacuation exit was also set.

[0057] For example, an obstacle model is also added to the scene model. In the obstacle model, the position, size, shape and height of the obstacles are close to the real environment.

[0058] For example, the scenario model also sets the number of characters and their initial positions. This is used to subsequently demonstrate the impact of the number of characters and their different locations on crowd evacuation. The characters include pedestrians and attackers.

[0059] In some embodiments, the set scene size is divided into multiple identical grids using a preset scale, with each grid being a cell.

[0060] For example, each character occupies one cell, while each obstacle can occupy multiple cells. Furthermore, both characters and obstacles have a certain height.

[0061] Step 101 constructs a scene model that closely resembles the real environment by setting up obstacle models and multiple roles, providing a reliable foundation for subsequent crowd evacuation simulation.

[0062] In step 102, a pedestrian model and an attacker model are constructed based on the scene model.

[0063] In some embodiments, a pedestrian model and an attacker model are constructed in the scene model built in step 101.

[0064] In some embodiments, constructing pedestrian and attacker models requires at least the construction of the figures' height, movement speed, and motion rules. The core content involves constructing the motion rules for pedestrians and attackers. The pedestrian motion rules describe how pedestrians move when facing dangerous situations requiring evacuation, while the attacker motion rules describe how attackers carry out their attack actions. This application provides detailed descriptions of these two types of motion rules.

[0065] First, constructing pedestrian movement rules can be done through steps 1021 to 1023, please refer to... Figure 2 The details of the pedestrian movement rules are as follows:

[0066] In step 1021, the pedestrian mole neighborhood, static ground field, and dynamic ground field are determined based on the scene model.

[0067] In some embodiments, in the constructed scene model, a pedestrian moves in space, and their location is surrounded by multiple cells. The pedestrian's Mole neighborhood is the set of the pedestrian's location and the eight adjacent cells. To accurately reflect the pedestrian's movement pattern, this application represents the pedestrian's movement direction in the Mole neighborhood, such as... Figure 3 As shown.

[0068] Reference Figure 3 A pedestrian can choose to remain stationary or move to any of their surroundings. The expression for the pedestrian's Moore neighborhood can be:

[0069]

[0070] In the formula, R i,j Let represent the set of the Mole neighborhood at cell (i, j), where i and j are the current positions of the pedestrian. and This represents the pedestrian's new position after they have moved.

[0071] In some embodiments, the static ground field is also defined in the constructed scene model to guide pedestrians to evacuation exits based on the static ground field value. The static ground field is determined by the shortest path to the evacuation exit; a larger static ground field value indicates a shorter path to the evacuation exit, making pedestrian evacuation easier. In the simulation, the pedestrian model automatically seeks the maximum static ground field value for evacuation. Therefore, the static ground field value is set to the maximum value at the evacuation exit location, and the value decreases as the distance from the evacuation exit increases.

[0072] Optionally, this application considers the impact of obstacles on pedestrian evacuation. When pedestrians encounter obstacles while moving towards an evacuation exit, they need to detour, and pedestrians will actively choose the shortest straight path. Therefore, in order to enable pedestrians to actively detour around obstacles during the simulation, the static ground field at the cell position occupied by the obstacle is set to a minimum value.

[0073] For example, after comprehensively considering the locations of evacuation exits and obstacles, the expression for the static ground field can be:

[0074] S ij =S e -L(x e ,y e ,x ij ,y ij )

[0075] In the formula, S ij S represents the static ground field value at the pedestrian's position (i,j). e This represents the static ground field value at evacuation exit location e, which is also the maximum static ground field value, x. e and y e The x-coordinate represents the location coordinates of evacuation exit e. ij and y ij L(x) represents the coordinates of the pedestrian at position (i,j). e ,y e ,x ij ,y ij ) represents the shortest path for a pedestrian to move to evacuation exit e.

[0076] It should be noted that pedestrians may need to navigate around multiple obstacles, therefore L(x) e ,y e ,xij ,y ij A path can be a single straight line or a combination of multiple straight lines.

[0077] It should also be noted that since there are multiple methods for calculating the shortest path, this application will not further limit each calculation method. For example, this application uses the A-star algorithm to calculate the shortest path.

[0078] In some embodiments, the dynamic field is also defined in the constructed scene model to keep pedestrians as far away from the attacker as possible. The so-called dynamic field is determined by the shortest path calculated based on the attacker's location. The larger the dynamic field value, the farther the pedestrian is from the attacker, and the easier it is for pedestrians to evacuate. In the simulation, the pedestrian model will automatically find the maximum value of the dynamic field for evacuation.

[0079] Optionally, this application considers the impact of obstacles on pedestrian evacuation. As pedestrians move towards the evacuation exit, they must not only detour around obstacles but also avoid attackers as much as possible. Therefore, the dynamic field value at the cell occupied by the obstacle is set to a minimum, and the dynamic field value at the cell occupied by the attacker is also set to a minimum to guide pedestrian evacuation.

[0080] For example, after comprehensively considering the attacker's position and the obstacle's position, the expression for the dynamic field can be:

[0081] D ij =L(x a y a x ij y ij )

[0082] In the formula, D ij x represents the dynamic ground field value of a pedestrian at position (i, j). a and y a The x represents the location coordinates of attacker 'a'. ij and y ij L(x) represents the coordinates of the pedestrian at position (i, j). a y a x ij y ij ) represents the shortest path from the attacker's location to the pedestrian's location.

[0083] It should be noted that pedestrians may need to navigate around multiple obstacles, therefore L(x) a y a x ij y ij A path can be a single straight line or a combination of multiple straight lines.

[0084] It should be noted that since there are multiple methods for calculating the shortest path, this application will not further limit each calculation method. For example, this application uses the A-star algorithm to calculate the shortest path.

[0085] In step 1022, the transition probability of a pedestrian in the Mole neighborhood is determined based on the pedestrian's Mole neighborhood, the static ground field, and the dynamic ground field.

[0086] In some embodiments, when a pedestrian receives a threat of attack, they will move to a relatively safe location in the surrounding area based on their own judgment. The relatively safe location is determined by the static and dynamic ground field values ​​of the pedestrian.

[0087] For example, the transition probability of a pedestrian in the Moore neighborhood can be expressed as:

[0088] P ij =N[exp(k s S ij +k d D ij (1-n) ij )]

[0089] In the formula, P ij S represents the transition probability of a pedestrian in cell (i, j); N represents the normalization factor, N∈[0,1]; ij k represents the static ground field value when a pedestrian is in cell (i, j); s k represents the sensitivity coefficient of the static ground field in the scene model. s ∈[0,1];D ij k represents the dynamic ground field value of a pedestrian in cell (i, j); d k represents the sensitivity coefficient of the dynamic ground field in the scene mode. d ∈[0,1]; n ij Indicates a placeholder, n ij It is 0 or 1 when n ij =0 indicates that cell (i, j) is occupied, and the pedestrian cannot move to this cell position. ij =1 indicates that cell (i, j) is not occupied and pedestrians can move to this cell position.

[0090] Furthermore, k s and k d As a sensitivity coefficient in scene mode, it can also be understood as a weight. When the sensitivity coefficient is 0, it means that the pedestrian's movement is not constrained by the static and dynamic ground fields, and the pedestrian moves randomly. When the sensitivity coefficient is 1, it means that the pedestrian's movement is completely constrained by the static and dynamic ground fields.

[0091] The purpose of adding a sensitivity coefficient in this application is to better reflect the actual situation. If there are many attackers and they are armed with powerful tools, the pedestrians' willingness to evacuate will be stronger, and they will be significantly affected by the static and dynamic field. Conversely, if there are few attackers and the harm is relatively small, the pedestrians' willingness to evacuate may be relatively weak, and they will be less affected by the static and dynamic field.

[0092] In step 1023, the moving target location of the pedestrian is obtained based on the transition probability.

[0093] In some embodiments, the pedestrian will move within the mole neighborhood based on the transition probability calculated in step 1022, and the target position of the transfer is the cell position with the highest transition probability value.

[0094] In step 1024, a conflict resolution solution for the transfer target location is set.

[0095] In some embodiments, multiple pedestrians may move towards the same destination during simultaneous evacuation, a situation that is quite common in real-world scenarios. Therefore, to address this, a solution needs to be defined for situations where multiple pedestrians have the same destination and a conflict occurs.

[0096] For example, this application stipulates that when multiple pedestrians have the same destination, the pedestrian with the highest probability of moving can move to the destination, while the other pedestrians will move to other locations with the highest probability of moving, other than the destination. That is, when a pedestrian finds that their destination is occupied, the pedestrian will move to the location that is easiest to evacuate from, other than the destination.

[0097] Steps 1021 to 1024 construct a pedestrian model based on the scene model, reflecting important factors such as the pedestrians' willingness to evacuate in real-world scenarios and the impact of obstacles on pedestrian evacuation. This pedestrian model treats pedestrians as intelligent agents with self-organizing and adaptive capabilities, and takes into account the individual characteristics of pedestrians and the interaction between pedestrians and the environment, thus making sufficient theoretical preparations to ensure the accuracy of subsequent simulation results.

[0098] Secondly, constructing the attacker's movement rules can be done through steps 1025 to 1027, as shown in the reference. Figure 4 The construction of the attacker's movement rules is detailed below:

[0099] In step 1025, the attacker's field of vision is determined based on the obstacle model and the attacker's field of vision model.

[0100] In some embodiments, the attacker first locates the pedestrian's position before launching an attack. However, the attacker's field of vision is not 360 degrees; they cannot see pedestrians to their sides or behind. Furthermore, multiple obstacles in the scene model also affect the attacker's field of vision. For example, if an obstacle is very high, even if a pedestrian is very close to the attacker, the attacker may not be able to see them due to the obstruction, thus preventing the attacker from attacking the pedestrian.

[0101] Optionally, this application introduces an attacker's field of vision model. In the attacker's field of vision model, a ray is first emitted directly in front of the attacker, and then the target hit by the ray is detected. If the ray hits a pedestrian, it means that the pedestrian is within the attacker's field of vision. If the ray hits an obstacle, it means that the attacker can only see the obstacle and cannot see the pedestrian behind the obstacle. At the same time, the attacker's field of vision angle needs to be preset so that he can only see a fan-shaped area.

[0102] Optionally, if no pedestrian is hit by the ray in the attacker's field of view model, the attacker can rotate in place to locate the pedestrian by adjusting the attacker's field of view until at least one pedestrian position appears in the attacker's field of view.

[0103] Therefore, based on the obstacle model and the attacker's field of vision model, the attacker's field of vision is determined as a fan-shaped area at a preset angle directly in front when there are no obstacles blocking it.

[0104] In step 1026, the attractiveness of the pedestrian to the attacker is determined based on the attacker's field of vision, attacker's location, and pedestrian's location.

[0105] In some embodiments, during the process of locating pedestrians, each pedestrian within the attacker's field of vision is attractive to the attacker. This application uses the attraction of pedestrians to the attacker to drive the attacker's attack behavior.

[0106] For example, an expression for attraction could be:

[0107]

[0108] In the formula, F pa Let x represent the pedestrian's attractiveness to the attacker, k be a constant, and x be a constant. p and y p Let x be the position coordinates of pedestrian p. a and y a Let be the coordinates of attacker a. Pedestrian p is within attacker a's field of view.

[0109] As can be seen from the above formula, the attraction of a pedestrian to an attacker is inversely proportional to the straight-line distance between the pedestrian and the attacker. That is, the closer the straight-line distance between the pedestrian and the attacker, the greater the attraction, and the farther the straight-line distance, the smaller the attraction.

[0110] In step 1027, the attacker's trajectory is determined based on the maximum attractiveness of the pedestrian to the attacker.

[0111] In some embodiments, an attacker calculates the attractiveness of pedestrians within their field of vision, locates the pedestrian with the greatest attractiveness, and then moves toward that pedestrian's location.

[0112] It should be noted that the attacker moves towards the pedestrian in a straight line. For example, the attacker's movement trajectory follows the trajectory calculated by A-star.

[0113] In some embodiments, a trigger is also set in the attacker movement rules. When certain preset conditions are met, the trigger will be automatically turned on, and the attacker will begin to execute the attacker movement rules.

[0114] For example, a preset condition could be that when the location of the pedestrian and the location of the attacker are both less than a preset threshold, the attacker does not need to locate the pedestrian and can directly attack the pedestrian.

[0115] For example, a preset condition could be that when multiple pedestrians have the same maximum attraction to the attacker, the attacker will randomly locate the pedestrian with the greatest attraction and move toward that pedestrian.

[0116] Steps 1025 to 1027 construct the attacker's movement rules based on the scene model. This allows the attacker to observe surrounding pedestrians through their field of vision, calculate the attractiveness of each pedestrian within their field of vision, and target the pedestrian with the highest attractiveness. The attacker then moves towards the target pedestrian by circling obstacles, thereby achieving the goal of chasing the pedestrian. In summary, step 102 constructs the movement rules for both pedestrians and the attacker based on the scene model, treating both as intelligent agents with self-organizing and adaptive capabilities. It also considers the individual characteristics of both pedestrians and the attacker, as well as their interaction with the environment, ensuring the accuracy of subsequent simulation results.

[0117] In step 103, an interaction model is constructed based on the pedestrian model and the attacker model.

[0118] In some embodiments, an interaction model between pedestrians and attackers is constructed based on the pedestrian model and attacker model built in step 102.

[0119] For example, the interaction model includes multiple interaction modes, such as a search mode, a chase mode, and an attack mode. The following is a description of the three modes in the example:

[0120] The pattern search involves locating the nearest pedestrian to the attacker based on the attacker's movement rules. The pattern search ends once the attacker has been located.

[0121] The chase mode involves an attacker chasing a pedestrian based on attacker movement rules, while the pedestrian moves away from the attacker based on pedestrian movement rules. The chase mode ends when the attacker catches up with the pedestrian. In this mode, the attacker can update the chase path based on the changing positions of both the attacker and the pedestrian.

[0122] The attack pattern is as follows: after the attacker catches up with the pedestrian, the pedestrian enters the attacked state. At this time, both the attacker and the pedestrian remain stationary. After a preset time, the pedestrian dies, and the attack pattern ends after the pedestrian dies.

[0123] Optionally, the various interaction modes can be switched between each other based on the simulation time step. For example, after locating the pedestrian's position through the search mode, the attacker enters the chase mode. In the chase mode, the attacker chases the pedestrian. After catching up with the pedestrian, the attacker enters the attack mode. In the attack mode, after the pedestrian dies, the attacker re-enters the search mode.

[0124] Furthermore, multiple interaction modes switch between each other until all pedestrians die and / or escape from the scene model.

[0125] In some embodiments, the interaction model further includes, when the attacker is chasing the target pedestrian in the chase mode, updating the pedestrian position of the shortest path in real time based on the real-time changing attacker position and pedestrian position, and changing the target pedestrian according to the pedestrian position of the shortest path.

[0126] Step 103 constructs the interaction mode between pedestrians and attackers. By creatively introducing multi-role interaction, the attacker's behavior can be determined based on the attacker's perception, providing a realistic basis and scientific support for the subsequent realistic simulation of the movement of each role in a horror scenario.

[0127] In step 104, crowd evacuation simulation is performed based on the scene model, pedestrian model, attacker model, and interaction model.

[0128] In some embodiments, on a simulation platform, crowd evacuation simulation under terrorist attack scenarios is performed by calling the scene model, pedestrian model, attacker model, and interaction model constructed in this application.

[0129] The crowd evacuation method provided in this application, based on a three-dimensional terrain model, conducts an in-depth analysis of the relationship between pedestrians and attackers, and presents corresponding interaction strategies, including pedestrian movement rules, attacker movement rules, and interaction patterns. The pedestrians' objective is to safely escape through evacuation exits, and their movement is driven by the static terrain influenced by the exit's location and the dynamic terrain influenced by the attacker's threat. The attacker's objective is to attack as many pedestrians as possible, and their movement is driven by the attraction of surrounding pedestrians. Furthermore, a field-of-view model is added for the attacker, allowing them to more accurately select target pedestrians, making the simulation results of this method more consistent with real-world situations.

[0130] This application provides a crowd evacuation method that can more realistically simulate crowd evacuation in a three-dimensional terrorist attack scenario. It takes into account the movement and interaction of pedestrians and attackers, as well as the impact of different obstacles on crowd movement. It can provide data support and scientific basis for the construction and management of public places, as well as for counter-terrorism organizations, in order to reduce casualties in terrorist attacks.

[0131] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0132] To verify the feasibility and beneficial effects of the method proposed in this application, cellular automata were used to simulate various obstacle layouts and the initial number of aliens to simulate pedestrian evacuation under a terrorist attack scenario. While observing pedestrian movement, the final death toll and evacuation time were obtained. The specific simulation process is as follows:

[0133] Experiment 1: Comparing the effects of placing a single obstacle in front of the exit.

[0134] exist Figure 5 A comparative experiment was conducted in the three scenarios shown. Figure 5 (a) shows a scenario without obstacles, (b) shows a scenario with a single low obstacle blocking the exit, and (c) shows a scenario with a single high obstacle blocking the exit. To more accurately analyze the impact of obstacles on evacuation and reduce other factors affecting evacuation results, in each identical scenario, there is one attacker and multiple pedestrians. Red cylinders represent attackers, and green cylinders represent pedestrians. It is assumed that all pedestrians and attackers move at the same speed, 4 units per second. The attacker's initial position is in the center of the room, and other pedestrians are randomly distributed throughout the room. The initial number of pedestrians is increased from 30 to 300, with each increase of 30 people conducted multiple times. That is, 30 people, 60 people, 90 people, 120 people, and so on, until reaching 300 people. The evacuation results are observed as the initial number of pedestrians changes.

[0135] Simulations were conducted under three different scenarios with varying initial pedestrian numbers to obtain data on pedestrian number minus death toll and pedestrian number minus evacuation time, as detailed in Tables 1 and 2 below.

[0136] Table 1. Number of deaths in accessible, high-obstacle, and low-obstacle scenarios.

[0137]

[0138] Table 2. Number of people evacuated in accessible, high-obstacle, and low-obstacle scenarios.

[0139]

[0140]

[0141] A detailed analysis and summary of Tables 1 and 2 were conducted, and the results are as follows: Figure 6 The line graph shown.

[0142] Reference Figure 6 The line graph shows that the number of deaths gradually increases with the number of pedestrians in the room under three scenarios. This indicates that the more pedestrians in the room, the more deaths. This suggests that a higher density of pedestrians in the room makes it easier for crowding to occur when fleeing from an attacker, thus making them vulnerable to attack. When the number of pedestrians remains constant, two situations need to be analyzed: first, the impact of obstacle height on the number of deaths. Figure 6 It is clear that the discounted values ​​for high-obstacle scenarios and low-obstacle scenarios almost overlap, indicating that the height of the obstacle has little impact on the number of deaths; secondly, the impact of placing obstacles in front of the exit on the number of deaths... Figure 6 It is clear that the number of deaths in the crowd evacuation simulation is higher in the unobstructed scenario than in the obstructed scenario. Furthermore, the difference between the obstructed and unobstructed scenarios becomes more pronounced as the number of pedestrians increases. Therefore, placing an obstacle in front of the exit can help pedestrians avoid attackers, reducing the risk of pedestrians being attacked and resulting in casualties.

[0143] In conclusion, the terrorist attack simulation experiment conducted in an unobstructed scenario resulted in a higher death toll and a longer evacuation time compared to the other two scenarios, especially when the number of pedestrians in the room increased. This indicates that placing a single obstacle in front of the exit has a beneficial effect on crowd evacuation.

[0144] Experiment 2: The Impact of Obstacle Density on Crowd Evacuation

[0145] Experiment 2 constructed a 3D space with multiple obstacles, similar to a train station waiting hall. Firstly, because train station waiting halls are locations prone to terrorist attacks, constructing this scenario is more realistic. Secondly, this scenario contains many obstacles, such as rows of benches, and the placement of these obstacles is not fixed. Therefore, calculating the density of each obstacle in this scenario allows for a better study of their distribution. The scenario constructed in this paper is as follows... Figure 7 As shown, Figure 7 In (a) and (b), 8 and 16 low obstacles are placed in the same scene, respectively, with the same direction perpendicular to the direction leading to the exit, to simulate the benches in the waiting room.

[0146] The initial conditions of the scenario are the existence of one attacker and multiple pedestrians. The attacker's position is at the center of the ground. The pedestrians' initial positions are randomly distributed, scattered among obstacles or other passageways in the space. Figure 7 (a) There are 8 obstacles of the same size, shape, and height, represented by cuboids, with their height below the pedestrian's line of sight. Divide the ground into four equal parts, and place 2 obstacles in each part. Figure 7 The obstacle in (b) and Figure 7 The same as in (a), only the number and layout are different. The ground is still divided into 4 equal parts, and 4 obstacles are placed in each part, resulting in 16 identical obstacles in the scene.

[0147] Simulations were conducted to study the movement of people during a terrorist attack. Through multiple experiments, data on the number of pedestrians minus the number of deaths, and the number of pedestrians minus evacuation time were obtained, as detailed in Tables 3 and 4.

[0148] Table 3. Number of deaths in high-density obstacle and low-density obstacle scenarios.

[0149]

[0150] Table 4 Evacuation Time for High-Density and Low-Density Obstacle Scenarios

[0151]

[0152] Further analysis and summarization of Tables 3 and 4 yielded the following results: Figure 8 The line graph shown.

[0153] Reference Figure 8 As the number of pedestrians increases, the number of deaths in the simulated terrorist attack scenarios also gradually increases in both scenarios, indicating that increased pedestrian numbers can lead to pedestrian deaths due to congestion in both scenarios. Furthermore, with the same number of pedestrians, the number of deaths in the low-density obstacle scenario is often greater than the number of deaths in the high-density obstacle scenario.

[0154] Therefore, appropriately increasing the number of obstacles can create multiple passages, increasing the paths pedestrians can take to avoid attackers during evacuation and reducing the probability of death.

[0155] Meanwhile, during the observation of the crowd evacuation simulation process, a "detour" phenomenon was discovered, that is, when pedestrians are being chased by attackers, pedestrians go around the obstacles to avoid the attackers' pursuit and create a distance between themselves and the attackers.

[0156] Meanwhile, during the observation of crowd evacuation simulation, a "dispersion" phenomenon was also discovered, that is, when a group of pedestrians is being chased by an attacker, they are divided into several groups by obstacles and evacuated from several different channels to reduce congestion and help them avoid the attack.

[0157] In conclusion, it can be observed that in scenarios with obstacles, a higher density of obstacles leads to a lower pedestrian mortality rate during evacuation.

[0158] Experiment 3: The impact of obstacle orientation on crowd evacuation.

[0159] To reduce the mortality rate in terrorist attacks, this application also conducted an in-depth study of the spatial layout of obstacles. Besides the number and density of obstacles affecting evacuation, the orientation of obstacles is also an important factor worthy of study. This paper modifies the orientation of obstacles to construct, for example... Figure 9 In the two scenarios, obstacles oriented in the same direction as the exit are referred to as "horizontal obstacles." If the passage formed between obstacles is perpendicular to the exit direction, it is referred to as a "vertical obstacle." Experiments were conducted in both scenarios to obtain data on the number of deaths and evacuation time. See Tables 5 and 6.

[0160] Table 5. Number of fatalities in scenarios with high-density horizontal obstacles and low-density horizontal obstacles.

[0161]

[0162]

[0163] Table 6 Evacuation times for scenarios with high-density horizontal obstacles and low-density horizontal obstacles.

[0164]

[0165] Experimental results analysis as follows Figure 10 As shown, in both scenarios, the number of deaths gradually increases with the increase in the number of pedestrians. This indicates that deaths will occur due to crowd congestion in both vertical and horizontal obstacle scenarios. While there are a few scenarios where the number of deaths is the same for both vertical and horizontal obstacles at the initial pedestrian count, overall, the number of deaths is higher in the vertical obstacle scenario than in the horizontal obstacle scenario. This suggests that horizontal obstacles can help pedestrians avoid attackers, reducing the probability of pedestrian death and ensuring that more pedestrians can safely leave through the exit.

[0166] In conclusion, to reduce the death toll in terrorist attacks in public places, as many horizontal barriers as possible should be placed, meaning that the passageways formed between the barriers should be aligned with the exit directions.

[0167] Corresponding to the crowd evacuation simulation method described in the above embodiments, Figure 11 The diagram shows a structural block diagram of the crowd evacuation simulation device provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0168] See Figure 11 The crowd evacuation simulation device in this application embodiment may include: a scene model establishment module 201, a character model establishment module 202, an interaction model establishment module 203, and a simulation module 204.

[0169] Scene model building module 201 is used to build scene models.

[0170] The scene model is a three-dimensional scene model. Constructing the scene model includes: setting the scene environment, which includes scene size, scene height, evacuation exit locations, evacuation exit sizes, obstacle models, number of characters, and initial character positions. The obstacle model includes obstacle locations, shapes, and heights. The scene size is divided into multiple grids based on a preset scale, with each grid representing a cell.

[0171] The character model building module 202 is used to construct pedestrian and attacker models based on the scene model. The pedestrian model includes pedestrian movement rules, and the attacker model includes attacker movement rules.

[0172] Optionally, pedestrian movement rules are constructed based on the scene model, including:

[0173] Based on the scene model, the pedestrian Moore neighborhood, static ground field, and dynamic ground field are determined. The pedestrian Moore neighborhood is the set of the pedestrian position and the 8 cells adjacent to the pedestrian position. The static ground field is determined by the shortest path calculated based on the evacuation exit position and obstacle model. The larger the static ground field value, the easier it is for pedestrians to evacuate. The dynamic ground field is determined by the shortest path calculated based on the attacker position and obstacle model. The larger the dynamic ground field value, the farther away from the attacker position, and the easier it is for pedestrians to evacuate.

[0174] Based on the pedestrian's Moore neighborhood, static ground field, and dynamic ground field, the transition probability of a pedestrian in the Moore neighborhood is determined. The transition probability expression is: P ij =N[exp(k s S ij +k d D ij (1-n) ij In the formula, P ij S represents the transition probability of a pedestrian in cell (i, j), N represents the normalization factor, N∈[0,1], and S ij Let k represent the static ground field when a pedestrian is in cell (i, j). sk represents the sensitivity coefficient of the static ground field. s ∈[0,1],D ij Let k represent the dynamic ground field of a pedestrian in cell (i, j). d k represents the sensitivity coefficient of the dynamic ground field. d ∈[0,1],n ij Indicates a placeholder, n ij It is 0 or 1 when n ij =0 indicates that cell (i, j) is occupied, and the pedestrian cannot move to this cell. When n ij =1 indicates that cell (i, j) is not occupied and pedestrians can move to this cell.

[0175] The pedestrian's moving target position is obtained based on the transition probability. The cell with the maximum transition probability is the pedestrian's moving target position.

[0176] A solution is defined for conflicting destination locations. The solution is that if multiple pedestrians have the same destination location, the pedestrian with the highest probability of moving to that destination location moves there, while the other pedestrians move to the location with the highest probability of moving among all other locations besides the destination location.

[0177] Optionally, attacker movement rules can be constructed based on the scenario model, including:

[0178] The attacker's field of vision is determined based on the obstacle model and the attacker's field of vision model. The attacker's field of vision model is a fan-shaped area at a preset angle directly in front of the attacker when there are no obstacles obstructing the view.

[0179] Based on the attacker's field of vision, attacker's location, and pedestrian's location, the maximum attraction of the pedestrian to the attacker is determined. The expression for the maximum attraction is:

[0180]

[0181] In the formula, F pa The value represents the maximum attractiveness of a pedestrian to an attacker, k represents a constant, and x represents the maximum attractiveness of a pedestrian to an attacker. p and y p Let x be the position coordinates of pedestrian p. a and y a Let be the coordinates of attacker a, where pedestrian p is within attacker a's field of vision.

[0182] The attacker's movement rules are determined based on the maximum attraction of pedestrians to the attacker, and the attacker's movement trajectory follows the A-star algorithm.

[0183] The character model building module 202 is also used to set triggers in the attacker model. When preset conditions are met, the trigger switch is activated, and the attacker begins to execute the attacker movement rules.

[0184] The interaction model building module 203 is used to build an interaction model based on the pedestrian model and the attacker model. The interaction model includes multiple interaction modes, which can be converted into each other.

[0185] Optionally, the interaction model building module 203 includes multiple interaction modes, including a search mode, a chase mode, and an attack mode. In the search mode, the attacker locates the nearest pedestrian based on attacker movement rules; the search mode ends once the attacker determines the nearest pedestrian's location. In the chase mode, the attacker chases the pedestrian based on attacker movement rules, while the pedestrian moves away from the attacker based on pedestrian movement rules. In the chase mode, the attacker can update the chase path based on the real-time changing positions of both the attacker and the pedestrian; the chase mode ends once the attacker catches up with the pedestrian. In the attack mode, after the attacker catches up with the pedestrian, both remain stationary until the pedestrian dies after a preset time; the attack mode ends upon the pedestrian's death. The attacker locates the pedestrian's position using the search mode, then enters the chase mode; after catching up with the pedestrian using the chase mode, the attacker enters the attack mode; and after the attack mode ends, the attacker re-enters the search mode.

[0186] Optionally, in the interaction model establishment module 203, when the attacker chases the target pedestrian in the chase mode, the shortest path pedestrian position is updated in real time based on the real-time changing attacker position and the real-time changing pedestrian position, and the target pedestrian is changed based on the shortest path pedestrian position.

[0187] Simulation module 204 is used to simulate crowd evacuation based on scene model, pedestrian model, attacker model and interaction model.

[0188] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0189] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0190] This application also provides a terminal device, see [link to relevant documentation] Figure 12 The terminal 300 may include: at least one processor 310 and a memory 320, wherein the memory 320 stores a computer program 321 that can run on the at least one processor 310, and the processor 310 executes the computer program 321 to implement the steps in any of the above method embodiments, for example... Figure 1 Steps 101 to 104 in the illustrated embodiment. Alternatively, when the processor 310 executes the computer program 321, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 11 The functions of modules 201 to 204 are shown.

[0191] For example, computer program 321 may be divided into one or more modules / units, one or more of which are stored in memory 320 and executed by processor 310 to complete this application. The one or more modules / units may be a series of computer program segments capable of performing specific functions, which describe the execution process of the computer program in terminal device 300.

[0192] Those skilled in the art will understand that Figure 12 This is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than shown, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0193] The processor 310 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0194] The memory 320 can be an internal storage unit of the terminal device or an external storage device, such as a plug-in hard drive, a smart media card (SMC), a secure digital card (SD), or a flash card. The memory 320 is used to store the computer program and other programs and data required by the terminal device. The memory 320 can also be used to temporarily store data that has been output or will be output.

[0195] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0196] The crowd evacuation simulation method provided in this application can be applied to terminal devices such as computers, wearable devices, vehicle-mounted devices, tablets, laptops, netbooks, personal digital assistants (PDAs), augmented reality (AR) / virtual reality (VR) devices, and mobile phones. This application does not impose any restrictions on the specific type of terminal device.

[0197] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the various embodiments of the crowd evacuation simulation method described above.

[0198] This application provides a computer program product that, when run on a mobile terminal, enables the mobile terminal to execute the steps described in the various embodiments of the crowd evacuation simulation method.

[0199] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0200] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0201] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0202] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0203] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0204] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method of crowd evacuation simulation, characterized by, The method comprises the following steps: constructing a scene model; constructing a pedestrian model and an attacker model based on the scene model, the pedestrian model comprising a pedestrian movement rule, and the attacker model comprising an attacker movement rule; constructing an interaction model based on the pedestrian model and the attacker model, the interaction model comprising a plurality of interaction modes which can be converted into each other; conducting crowd evacuation simulation based on the scene model, the pedestrian model, the attacker model and the interaction model; the scene model is a three-dimensional space scene model, and the construction of the scene model comprises: setting a scene environment, the scene environment comprising a scene size, a scene height, an evacuation port position, an evacuation port size, an obstacle model, a number of persons and initial positions of the persons, wherein the obstacle model comprises an obstacle position, an obstacle shape and an obstacle height; dividing the scene size into a plurality of grids based on a preset scale, each grid being a cell; constructing the pedestrian movement rule based on the scene model, comprising: determining a pedestrian Moore neighborhood, a static field and a dynamic field based on the scene model, the pedestrian Moore neighborhood being a set of positions of pedestrians and eight cells adjacent to the positions of the pedestrians, the static field being determined based on a shortest path calculated based on the evacuation port position and the obstacle model, the greater the value of the static field, the easier the evacuation of the pedestrians, and the dynamic field being determined based on a shortest path calculated based on the attacker position and the obstacle model, the greater the value of the dynamic field, the farther the distance from the attacker position, and the easier the evacuation of the pedestrians; determining a transition probability of the pedestrians in the Moore neighborhood based on the pedestrian Moore neighborhood, the static field and the dynamic field, the expression of the transition probability being: In the formula, Indicates that pedestrians are in cells ( i , j The transition probability at time ) Represents the normalization factor. , Indicates that pedestrians are in cells ( i , j The static ground field value at that time. This represents the sensitivity coefficient of the static ground field in the scene model. , Indicates that pedestrians are in cells ( i , j The dynamic field value at that time. This represents the sensitivity coefficient of the dynamic ground field in the scene mode. , Indicates a placeholder. It is 0 or 1 when Indicates cell ( i , j () is occupied, and pedestrians cannot move to this cell position when Indicates cell ( i , j If the cell is not occupied, pedestrians can move to this cell location; obtaining a moving target position of the pedestrians based on the transition probability, the cell with the maximum value of the transition probability being the transition target position of the pedestrians; setting a conflict resolution scheme for the transition target position, the resolution scheme being that, if the transition target positions of a plurality of pedestrians are the same, the pedestrian with the maximum transition probability moves to the transition target position, and other pedestrians move to positions with the maximum transition probability other than the transition target position; constructing the attacker movement rule based on the scene model, comprising: determining an attacker field of view based on the obstacle model and an attacker field of view model, the attacker field of view model being a fan-shaped region in front of a preset angle without being blocked by obstacles; determining an attraction of the pedestrians to the attacker based on the attacker field of view, an attacker position and a pedestrian position, the expression of the attraction being: wherein, represents the attractiveness of the pedestrian to the attacker, represents a constant, and are position coordinates of the pedestrian p , and are position coordinates of the attacker a , wherein the pedestrian p is within the field of view of the attacker a . determining an attacker movement rule based on the maximum attraction of the pedestrians to the attacker, the movement trajectory of the attacker being a movement trajectory following an A-star algorithm.

2. The method of claim 1, wherein, the plurality of interaction modes comprise a seeking mode, a pursuit mode and an attack mode; the seeking mode is that the attacker locates a pedestrian closest to the attacker based on the attacker movement rule, and the seeking mode ends after the attacker determines the position of the closest pedestrian. The pursuit mode is that the attacker pursues the pedestrian based on the attacker motion rule, the pedestrian moves away from the attacker based on the pedestrian motion rule, and the attacker can update the pursuit path based on the real-time changing attacker position and real-time changing pedestrian position in the pursuit mode. The pursuit mode ends when the attacker catches up with the pedestrian; The attack mode is that the attacker and the pedestrian remain stationary after the attacker catches up with the pedestrian, and the pedestrian dies after a preset time. The attack mode ends after the pedestrian dies; The attacker enters the pursuit mode after locating the pedestrian position through the search mode, pursues the pedestrian in the pursuit mode, enters the attack mode after catching up with the pedestrian, and reenters the search mode after the pedestrian dies in the attack mode until all pedestrians die and / or escape from the scene model.

3. The method of claim 2, wherein, When the attacker pursues the target pedestrian in the pursuit mode, the pedestrian position of the shortest path is updated in real time based on the real-time changing attacker position and real-time changing pedestrian position, and the target pedestrian is changed based on the pedestrian position of the shortest path.

4. The method of claim 1, wherein, The trigger is arranged in the attacker model, and the trigger switch is triggered when a preset condition is met, and the attacker starts to execute the attacker motion rule.

5. A crowd evacuation simulation apparatus characterized by comprising: It comprises: a scene model establishing module for constructing a scene model; a person model establishing module for constructing a pedestrian model and an attacker model based on the scene model, wherein the pedestrian model comprises a pedestrian motion rule, and the attacker model comprises an attacker motion rule; an interaction model establishing module for constructing an interaction model based on the pedestrian model and the attacker model, wherein the interaction model comprises multiple interaction modes that can be converted to each other; a simulation module for conducting crowd evacuation simulation based on the scene model, the pedestrian model, the attacker model and the interaction model; The scene model is a three-dimensional scene model, and the construction of the scene model comprises: setting a scene environment, wherein the scene environment comprises scene size, scene height, evacuation port position, evacuation port size, obstacle model, number of persons and initial position of persons, the obstacle model comprises obstacle position, obstacle shape and obstacle height, and the scene size is evenly divided into multiple grids based on a preset scale, and each grid is a cell; The pedestrian motion rule is constructed based on the scene model, comprising: Based on the scene model, a pedestrian Moore neighborhood, a static field and a dynamic field are determined, the pedestrian Moore neighborhood is a set of a pedestrian position and eight cells adjacent to the pedestrian position, the static field is determined based on the shortest path calculated based on the evacuation port position and the obstacle model, the greater the static field value, the easier the pedestrian evacuation, and the dynamic field is determined based on the shortest path calculated based on the attacker position and the obstacle model, the greater the dynamic field value, the farther the distance from the attacker position, and the easier the pedestrian evacuation; Based on the pedestrian Moore neighborhood, the static field and the dynamic field, the transition probability of the pedestrian in the Moore neighborhood is determined, and the transition probability expression is: In the formula, Indicates that pedestrians are in cells ( i , j The transition probability at time ) Represents the normalization factor. , Indicates that pedestrians are in cells ( i , j The static ground field value at that time. This represents the sensitivity coefficient of the static ground field in the scene model. , Indicates that pedestrians are in cells ( i , j The dynamic field value at that time. This represents the sensitivity coefficient of the dynamic ground field in the scene mode. , Indicates a placeholder. It is 0 or 1 when Indicates cell ( i , j () is occupied, and pedestrians cannot move to this cell position when Indicates cell ( i , j If the cell is not occupied, pedestrians can move to this cell location; acquire a moving target position of the pedestrian based on the transition probability, a cell with a maximum value of the transition probability being the moving target position of the pedestrian; set a conflict resolution scheme for the moving target position, the resolution scheme being that if the moving target positions of multiple pedestrians are the same, a pedestrian with the maximum transition probability moves to the moving target position, and other pedestrians move to a position with the maximum transition probability other than the moving target position; construct the attacker motion rule based on the scenario model, including: determine an attacker field of view based on the obstacle model and an attacker field of view model, the attacker field of view model being a fan-shaped region with a preset angle in front without being blocked by an obstacle; determine an attraction of the pedestrian to the attacker based on the attacker field of view, an attacker position, and a pedestrian position, an expression of the attraction being: In the formula, This indicates the pedestrian's attractiveness to the attacker. Represent a constant. and For pedestrians p Location coordinates, and attacker a The location coordinates, where the pedestrian p Attacker a Within their field of vision; determine the attacker motion rule based on the maximum attraction of the pedestrian to the attacker, the attacker motion trajectory being a motion trajectory following an A-star algorithm. 6.A terminal device, comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and the computer program comprises the following steps of: The processor executes the computer program to implement the method in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: receiving a request for a resource from a client; determining whether the client is authorized to access the resource; and if the client is authorized to access the resource, providing the resource to the client. The computer program is executed by the processor to implement the method in any one of claims 1 to 4.

Images