Crowd evacuation simulation method, device, system, equipment, medium and product

Abstract

This disclosure provides a method, apparatus, system, device, medium, and product for simulating crowd evacuation. The method includes: acquiring scene data in the area to be evacuated; mapping the scene data to a spatial model of the area to be evacuated to determine the environmental impact parameters and neighboring individual impact parameters for each individual to be evacuated; determining the current emotional level of each individual to be evacuated from the environmental impact parameters and neighboring individual impact parameters using a first emotion-driven model; determining the behavioral pattern of each individual to be evacuated based on the current emotional level; and searching for the evacuation path of each individual to be evacuated from the spatial model using a path planning algorithm. During the simulation of crowd evacuation, the method can more accurately learn the impact of individual emotions on crowd evacuation, improving simulation accuracy. A comprehensive cost function is constructed based on the coupled parameters of emotional level and behavioral pattern, and path planning is performed to provide more accurate path guidance for individuals to be evacuated in the area to be evacuated.

Description

Technical Field

[0001] This disclosure relates to the field of security technology, and in particular to a method, apparatus, system, equipment, medium and product for simulating crowd evacuation. Background Technology

[0002] During emergency evacuation in enclosed spaces, individual decisions and behaviors are not only influenced by the physical environment (such as the location of exits and the distribution of obstacles), but are also easily dominated by the spread and diffusion of panic. In emergency situations (such as fires, earthquakes, gas leaks, etc.), like infectious diseases, panic can spread rapidly among people through visual, auditory, and physical contact, leading to irrational behaviors such as blindly following others (herd effect), rushing for narrow passages or single exits (exit focus), ignoring safety warnings and going against the flow (risk misjudgment), and even extreme situations such as pushing and trampling due to loss of emotional control, seriously hindering evacuation efficiency and threatening lives.

[0003] Existing crowd evacuation technologies use simulation methods to optimize evacuation plans, usually aiming to rationally choose the shortest path. However, they generally lack emotional factors and cannot truly reflect the impact of panic and other emotions on individuals in an emergency, resulting in a large deviation between simulation results and the actual situation on site. Summary of the Invention

[0004] The technical problem to be solved by this disclosure is to overcome the lack of emotional factors in the evacuation schemes of the prior art, and to provide a crowd evacuation simulation method, device, system, equipment, medium and product.

[0005] This disclosure solves the above-mentioned technical problems through the following technical solution:

[0006] In a first aspect, embodiments of this disclosure provide a crowd evacuation simulation method, the method comprising:

[0007] Acquire scene data in the area to be evacuated; the scene data includes sensor data of hazardous materials and behavioral data of each individual to be evacuated in the crowd;

[0008] The scene data is mapped to the spatial model of the area to be evacuated in order to determine the environmental impact parameters and the impact parameters of neighboring individuals for each individual to be evacuated.

[0009] The current emotional level of each individual to be evacuated is determined using a first emotion-driven model based on environmental impact parameters and neighborhood individual impact parameters.

[0010] Determine the behavioral pattern of each individual to be evacuated based on the current emotional level;

[0011] Based on the behavioral patterns, a comprehensive cost function is established for each individual to be evacuated, and a path planning algorithm is used to search for an evacuation path for each individual to be evacuated from the spatial model.

[0012] Optionally, the scene data may also include map data of the area to be evacuated;

[0013] The method further includes:

[0014] A spatial model is established based on the map data; the spatial model includes at least a grid set, a node set, and an edge set. Each grid cell in the grid set represents a passable or impassable sub-region in the area to be evacuated. Each node in the node set represents a preset sub-region in the area to be evacuated. Each edge in the edge set represents a passageway between two preset sub-regions.

[0015] Optionally, the sensing data includes the concentration of hazardous materials;

[0016] The step of mapping the scene data to the spatial model of the area to be evacuated, in order to determine the environmental impact parameters of each individual to be evacuated, includes:

[0017] Map the hazardous materials and each individual to be evacuated to the spatial model to determine the spatial location of the hazardous materials and each individual to be evacuated;

[0018] The concentration of the hazardous materials and the spatial location of the hazardous materials and each individual to be evacuated are calculated based on the hazardous field model to obtain the environmental impact parameters of each individual to be evacuated.

[0019] Optionally, the behavioral data includes the historical emotional level of each of the individuals to be evacuated;

[0020] The step of mapping the scene data to the spatial model of the area to be evacuated, in order to determine the influence parameters of neighboring individuals for each individual to be evacuated, includes:

[0021] Each individual to be evacuated is mapped into the spatial model to determine the spatial location of each individual to be evacuated;

[0022] Determine the neighboring individuals to be evacuated within a preset distance range for each individual to be evacuated;

[0023] Based on the distance between the neighboring individuals to be evacuated and each of the individuals to be evacuated, the emotional levels of the neighboring individuals to be evacuated are integrated and averaged to obtain the influence parameters of the neighboring individuals for each individual to be evacuated.

[0024] Optionally, the scene data may also include environmental guidance data;

[0025] The method further includes:

[0026] The environmental guidance data is used to determine the reassurance effect parameters on each individual to be evacuated based on a preset relationship.

[0027] The step of determining the current emotional level of each individual to be evacuated from environmental influence parameters and neighboring individual influence parameters using the first emotion-driven model includes:

[0028] The first emotion-driven model determines the current emotional level of each individual to be evacuated from environmental influence parameters, neighborhood individual influence parameters, and reassurance influence parameters.

[0029] Optionally, the first emotion-driven model includes:

[0030] ;

[0031] Where t represents a historical moment. For the current moment, Let k be the current emotional level of the k-th individual to be evacuated at the current moment. For the influence parameters of neighboring individuals, For environmental impact parameters, To appease the affected parameters, The emotional contagion coefficient, The environmental stimulus coefficient, To intervene and appease the coefficient, The emotional self-regulation coefficient. , and All are constants.

[0032] Optionally, the behavioral data may also include the physical compressive force of each of the individuals to be evacuated;

[0033] The method further includes:

[0034] When the physical pressure on the individual to be evacuated is greater than or equal to the first preset pressure and less than the second preset pressure, the current emotional level of the neighboring individuals to be evacuated within a preset distance range of the individual to be evacuated is increased.

[0035] When the physical pressure on the individual to be evacuated is greater than or equal to the second preset pressure, an extreme risk warning is triggered, and the individual to be evacuated is transformed into an obstacle node in the spatial model, so as to replan the evacuation path of the neighboring individuals to be evacuated within a preset distance range of the individual to be evacuated.

[0036] Optionally, the behavioral data may also include the historical emotional level of each of the individuals to be evacuated;

[0037] The method further includes:

[0038] If the duration of the historical emotional level of the individual to be evacuated being greater than or equal to the first preset level is greater than or equal to the preset duration, the current emotional level of the individual to be evacuated is determined by the second emotion-driven model; the second emotion-driven model obtains the current emotional level by attenuating the historical emotional level with preset attenuation parameters and preset duration.

[0039] Optionally, determining the behavioral pattern of each individual to be evacuated based on the current emotional level includes:

[0040] When the current emotional level is greater than or equal to the first preset level, the corresponding individual to be evacuated is determined to be in an irrational behavior pattern.

[0041] Alternatively, if the current emotional level is greater than a first preset level, the corresponding individual to be evacuated is determined to be in a rational behavior pattern.

[0042] Optionally, the comprehensive cost function includes:

[0043] ;

[0044] in, For each of the individuals to be evacuated to the exit The shortest path distance, For access to the said exit The degree of congestion of the path, The path hazard level, For path recommendation score, For following coefficient, To select the exit The proportion of the individuals to be evacuated to all the individuals to be evacuated. Distance weights For crowding weight, Risk weight, To guide weights, To follow the weight, , , , All are constants.

[0045] Optionally, establishing a comprehensive cost function for each individual to be evacuated based on the behavioral pattern includes:

[0046] When the behavior pattern of the individuals to be evacuated is a rational behavior pattern, the distance weight in the corresponding comprehensive cost function is reduced, and the crowding weight and guidance weight in the corresponding comprehensive cost function are increased.

[0047] Alternatively, when the behavior pattern of the individuals to be evacuated is an irrational behavior pattern, the distance weight in the corresponding comprehensive cost function is increased, and the crowding weight and guidance weight in the corresponding comprehensive cost function are decreased.

[0048] Optionally, after establishing a comprehensive cost function for each individual to be evacuated based on the behavioral pattern, and searching for an evacuation path for each individual to be evacuated from the spatial model using a path planning algorithm, the process includes:

[0049] The motion state of each individual to be evacuated is determined based on the micro-motion model, and the spatial position of each individual to be evacuated is updated based on the motion state.

[0050] Determine whether each individual to be evacuated has left the evacuation area based on the updated spatial location;

[0051] If the determination is negative, continue with the step of obtaining scene data in the area to be evacuated until each individual to be evacuated leaves the area to be evacuated.

[0052] Optionally, the micro-motion model includes a social force model;

[0053] Determining the motion state of each individual to be evacuated based on the micro-motion model includes:

[0054] The current emotional level of each individual to be evacuated is mapped to multiple moderating factors in the social force model to determine the movement state of each individual to be evacuated; the moderating factors include self-driving force, interpersonal interaction force, boundary force, and emotional driving additional force.

[0055] Optionally, the emotion-driven additional force includes:

[0056] ;

[0057] in, For individuals awaiting evacuation The added force of emotion-driven action Emotion-driven weights Let be the unit vector in the export direction. For individuals awaiting evacuation speed, For individuals awaiting evacuation speed, For the preset distance range, For individuals awaiting evacuation with individuals to be evacuated The distance between them.

[0058] Secondly, embodiments of this disclosure provide a crowd evacuation simulation device, the device comprising:

[0059] The acquisition module is used to acquire scene data in the area to be evacuated; the scene data includes sensor data of hazardous materials and behavioral data of each individual to be evacuated in the crowd;

[0060] The mapping module is used to map the scene data to the spatial model of the area to be evacuated, so as to determine the environmental impact parameters and the impact parameters of neighboring individuals for each individual to be evacuated.

[0061] The first level determination module is used to determine the current emotional level of each individual to be evacuated from environmental influence parameters and neighborhood individual influence parameters using a first emotion-driven model.

[0062] The pattern determination module is used to determine the behavioral pattern of each individual to be evacuated based on the current emotional level.

[0063] The planning module is used to establish a comprehensive cost function for each individual to be evacuated based on the behavioral pattern, and to search for an evacuation path for each individual to be evacuated from the spatial model in combination with a path planning algorithm.

[0064] Thirdly, embodiments of this disclosure provide a crowd evacuation simulation system, the system comprising a data acquisition layer, a data processing layer, and an evacuation layer;

[0065] Both the data acquisition layer and the evacuation layer are connected to the data processing layer. The data acquisition layer is used to collect scene data. The data processing layer is loaded with the crowd evacuation simulation method as described in any one of the first aspects. The data processing layer is used to determine the evacuation path based on the scene data. The evacuation layer is used to evacuate the crowd based on the evacuation path.

[0066] Fourthly, embodiments of this disclosure provide an electronic device, including a memory, a processor, and a computer program stored in the memory and for running on the processor, wherein the processor executes the computer program to implement the crowd evacuation simulation method as described in any one of the first aspects.

[0067] Fifthly, embodiments of this disclosure provide a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the crowd evacuation simulation method as described in any one of the first aspects.

[0068] In a sixth aspect, embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, implements the crowd evacuation simulation method as described in any one of the first aspects.

[0069] The positive and progressive effects of this disclosure are as follows:

[0070] This disclosure maps sensor data of hazardous materials and behavioral data of neighboring individuals into a spatial model to simulate the impact of the environment on the emotional level of individuals to be evacuated. It then establishes coupling parameters between emotional level and behavioral patterns. In the process of simulating crowd evacuation, it can more accurately learn the impact of individual emotions on crowd evacuation, improve simulation accuracy, and construct a comprehensive cost function based on the coupling parameters of emotional level and behavioral patterns to perform path planning, providing more accurate path guidance for individuals to be evacuated in the evacuation area.

[0071] 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 disclosure. Attached Figure Description

[0072] Figure 1 A first flowchart illustrating a crowd evacuation simulation method provided as an exemplary embodiment of this disclosure;

[0073] Figure 2 A first flowchart illustrating step S102 of a crowd evacuation simulation method provided as an exemplary embodiment of this disclosure;

[0074] Figure 3 A second flowchart illustrating step S102 of a crowd evacuation simulation method provided as an exemplary embodiment of this disclosure;

[0075] Figure 4 This disclosure provides a second flowchart illustrating an exemplary embodiment of a crowd evacuation simulation method.

[0076] Figure 5 This disclosure provides a schematic diagram of the third process of an exemplary embodiment of a crowd evacuation simulation method;

[0077] Figure 6 A schematic diagram of a crowd evacuation simulation device provided as an exemplary embodiment of this disclosure;

[0078] Figure 7 A schematic diagram of a crowd evacuation simulation system provided as an exemplary embodiment of this disclosure;

[0079] Figure 8 A schematic diagram of the structure of an electronic device provided for an exemplary embodiment of this disclosure. Detailed Implementation

[0080] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0081] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0082] In this embodiment of the disclosure, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good morals.

[0083] The following describes a crowd evacuation simulation method provided by an embodiment of this disclosure. Figure 1 This is a flowchart illustrating a crowd evacuation simulation method provided in this disclosure. This specification provides the method operation steps as shown in the embodiments or flowcharts, but based on conventional or non-inventive methods, more or fewer operation steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual system or server product execution, the method can be executed sequentially according to the embodiments or drawings, or in parallel (e.g., in scenarios involving parallel processors or multi-threaded processing). Specifically, as shown... Figure 1 As shown, the method may include:

[0084] S101. Obtain scene data in the area to be evacuated.

[0085] The scene data includes, but is not limited to, sensor data of hazardous materials and behavioral data of each individual to be evacuated within the crowd. In some cases, scene data also includes map data and environmental guidance data of the evacuation area. Sensor data of hazardous materials can be collected through smoke sensors, temperature sensors, etc. This data is used to measure the degree of hazard of hazardous materials in the evacuation area, thus reflecting the impact of hazardous materials on the behavior of each individual to be evacuated. Behavioral data of each individual to be evacuated can be collected through cameras. This data reflects the state of each individual during the evacuation process and also reflects the impact of each individual on neighboring individuals to be evacuated.

[0086] In one embodiment, the spatial model of the evacuation area is described, and the scene data also includes map data of the evacuation area. The specific map data can also be obtained by scanning the space based on lidar or image sensors deployed in the evacuation area. The method also includes: establishing a spatial model based on the map data.

[0087] The spatial model includes at least a grid set, a node set, and an edge set. Each grid cell in the grid set represents a passable or impassable sub-region within the evacuation area. Each node in the node set represents a predefined sub-region within the evacuation area. Each edge in the edge set represents a passageway between two predefined sub-regions. Specifically, the spatial discretization and topological structure of the spatial model are as follows:

[0088] First, grid representation. The area to be evacuated is divided into regular or irregular discrete grids, resulting in a grid set. Each grid cell represents a passable or impassable area. Each grid cell has the following attributes: geometric attributes include area, center coordinates, and connectivity with neighboring grid cells; environmental attributes include sensor data of hazardous materials, such as smoke concentration, temperature, and toxic gas concentration; and functional attributes include whether it is an exit cell, an obstacle, or a main passageway.

[0089] Secondly, graph structure representation. Pre-defined sub-regions within the evacuation area (e.g., rooms, corridors, stairwells, exits, etc.) are abstracted as sets of nodes. The passage between two predefined sub-regions is abstracted as a set of edges. For each edge Set capacity ,length and congestion coefficient .

[0090] Third, a hybrid representation of mesh and graph structures is adopted. A hybrid modeling approach of "mesh + graph" is used, with fine meshes used locally to simulate microscopic crowd flow, and a graph structure used overall for path search and exit allocation, balancing simulation accuracy and computational efficiency, consistent with the logic of adaptive modeling for complex regions.

[0091] During crowd evacuation simulation, the inventors discovered that existing crowd evacuation simulations often simplify the evacuation area into a regular rectangle, modeled using a two-dimensional Cartesian coordinate system. This method ignores the complex structure of "irregular areas (such as circular exhibition halls), areas with multiple obstacles (such as shopping mall shelving), and areas with multiple exits (such as subway transfer passages)" in actual enclosed spaces. This simplification leads to insufficient accuracy in evacuation route planning and risk warning; for example, it cannot simulate the risk of congestion caused by panic in narrow passages between shelves. The spatial model of the evacuation area constructed in this disclosure, based on the actual structural characteristics of the enclosed space, automatically collects area map data (including but not limited to irregular area boundaries, spatial location and size of obstacles, and distribution of multiple exits) through a fusion of "LiDAR scanning + camera image recognition," constructing a regular or irregular discrete grid that fits the actual evacuation scenario, replacing traditional rectangular area modeling. For example, in a shopping mall scenario, the system can automatically identify narrow passage sub-areas formed by shelving and pedestrian confluence sub-areas formed by elevator entrances, accurately recreating them in the spatial model and avoiding risk misjudgment caused by area simplification.

[0092] S102. Map the scene data to the spatial model of the area to be evacuated in order to determine the environmental impact parameters of each individual to be evacuated and the impact parameters of neighboring individuals.

[0093] Specifically, the sensor data of hazardous materials is mapped onto the spatial model of the area to be evacuated to obtain the environmental impact parameters for each individual to be evacuated. These environmental impact parameters reflect the impact of hazardous materials in the area to be evacuated on the emotional level of the individual to be evacuated. The behavioral data is mapped onto the spatial model of the area to be evacuated to obtain the neighborhood individual impact parameters for each individual to be evacuated. These neighborhood individual impact parameters reflect the impact of the emotions or behaviors of other individuals to be evacuated around each individual on the emotional level of that individual.

[0094] In one embodiment, the evacuation scenario typically involves a danger caused by fire, gas leak, or liquid leak in the area to be evacuated, which leads to the need for the evacuation of people. Therefore, the sensor data of hazardous materials includes the concentration of hazardous materials, such as liquid or gas concentration, while the behavioral data of each individual to be evacuated includes, but is not limited to, spatial location and movement status.

[0095] Specifically, sensors (such as infrared counters, cameras, and mobile terminal positioning) are used to collect real-time behavioral data on the number of individuals to be evacuated, the spatial location and movement of each individual, and the concentration of hazardous materials. This data provides input to the spatial model, ensuring a high degree of consistency between the simulation and the actual situation on site. Additionally, the behavioral data of the individuals to be evacuated may include the emotional level calculated by the first emotion-driven model. Generally, the initial emotional level is usually 0. During subsequent evacuation and simulation processes, the behavioral data and emotional level will be updated using the simulation results of path planning and behavioral patterns, which will be detailed later.

[0096] In an alternative implementation, see Figure 2 Step S102 specifically includes:

[0097] S1021. Map the hazardous materials and each individual to be evacuated into the spatial model to determine the spatial location of the hazardous materials and each individual to be evacuated.

[0098] For each individual to be evacuated, the mapping in the spatial model is specifically defined as follows: Each individual to be evacuated The state variables in the spatial model are as follows: Spatial location: at each time step. The next grid or node Motion status: Current speed Maximum achievable speed Target direction Mood level Cognitive and decision-making parameters: path planning weights (shortest distance, safety, density penalty), and compliance with broadcast / guidance. Social connectivity and follower behavior weights, etc.

[0099] For the mapping of hazardous materials in the spatial model, the corresponding state variables are as follows: Spatial location: the grid or node it is located at each time step. Concentration of hazardous materials wait.

[0100] S1022. Based on the hazardous field model, calculate the concentration of hazardous materials and the spatial location of hazardous materials and each individual to be evacuated to obtain the environmental impact parameters of each individual to be evacuated.

[0101] The essence of the hazard field model lies in its representation of environmental risk as a continuous field, coupled with the individuals to be evacuated. The key lies in two relationships: how the hazard field itself evolves, and how the individuals to be evacuated perceive and respond to it. Specifically, this is determined by the hazard field model. The following is a standardized description of hazardous materials such as fire, smoke, and toxic gases:

[0102] First, a simulation using diffusion equations describes the spatiotemporal evolution of hazardous materials in space:

[0103] ;

[0104] in, For each moment Location of hazardous materials , The concentration of hazardous materials, Where is the diffusion coefficient. This is the attenuation coefficient.

[0105] Next, a discrete difference scheme is used to solve the diffusion equation in a linked manner, so as to obtain the time-series model at each moment. Each grid Concentration on Further construct the hazard field model for each grid:

[0106] ;

[0107] in, For each moment The spatial location of each grid cell. It can be a linear mapping or a nonlinear function with a threshold, substituted with the spatial location of the individual to be evacuated. This yields the danger index of the spatial location of the individuals to be evacuated. .

[0108] Next, environmental impact parameters are calculated based on the hazard index of the spatial location of the individuals to be evacuated, in order to quantify the impact of hazardous materials in the evacuation area on the individuals to be evacuated. The formula is as follows:

[0109] ;

[0110] in, For environmental impact parameters, The danger index of the spatial location of the individuals to be evacuated. The sensitivity level of the individuals to be evacuated can be configured according to their characteristics. For example, the value is 1.2-1.5 for the elderly and children, and 1.0 for the general population.

[0111] Additionally, the spatial location of the exit or passageway can be determined. Substituting into the hazard field model, the grid hazard index of the exit or passage is obtained. ,when If the value is greater than or equal to a preset threshold, it is determined that the passage is not allowed. The preset threshold is preferably 0.8, but it can be configured according to the actual situation.

[0112] In an alternative implementation, see Figure 3 Step S102 specifically includes:

[0113] S102-1. Map each individual to be evacuated to the spatial model and determine the spatial location of each individual to be evacuated.

[0114] For each individual to be evacuated, the mapping in the spatial model is specifically defined as follows: Each individual to be evacuated The state variables in the spatial model are as follows: Spatial location: the current grid or node. Motion status: Current speed Maximum achievable speed Target direction Mood level Cognitive and decision-making parameters: path planning weights (shortest distance, safety, density penalty), and compliance with broadcast / guidance. Social connectivity and follower behavior weights, etc.

[0115] S102-2. Determine the neighboring individuals to be evacuated within a preset distance range for each individual to be evacuated.

[0116] The preset distance range is defined as the neighborhood of each individual to be evacuated. Generally, the preset distance range is set to a radius of 5 meters from the spatial location of each individual to be evacuated, but it is not limited to this and can be adjusted according to actual needs.

[0117] S102-3. Based on the distance between the neighboring individuals to be evacuated and each other, the emotional level of the neighboring individuals to be evacuated is integrated and averaged to obtain the neighboring individual influence parameter for each individual to be evacuated.

[0118] Determining the influence parameters of neighboring individuals essentially characterizes the interaction mechanism between group behavior and emotional levels. In real evacuation scenarios, humans are strongly coupled systems, and emotions have significant contagiousness and amplification effects. The purpose of determining the influence parameters of neighboring individuals is threefold: First, to simulate the emotional contagion between individuals in real-world scenarios. In high-risk environments, people perceive the behavior of those around them (running, shouting, crowding) through sight and hearing and quickly adjust their own state. Second, to influence individuals' path planning. The more panicked the individuals in the neighborhood are, the more likely they are to abandon rational path planning and follow the crowd. Third, to avoid the risk of congestion and stampedes in the evacuation area. The more panicked the individuals in the neighborhood are, the smaller the safe distance between them becomes, making it easier for them to gather at the same exit.

[0119] Specifically, the emotion level is calculated based on the first emotion-driven model. The initial emotion level is usually 0. The formula for the influence parameter of neighboring individuals is as follows:

[0120] ;

[0121] in, For the influence parameters of neighboring individuals, For individuals in the surrounding area awaiting evacuation, For the preset distance range, For individuals in the neighborhood awaiting evacuation emotional level For individuals awaiting evacuation emotional level Neighborhood individuals awaiting evacuation with individuals to be evacuated The distance between them.

[0122] In addition, in step S102-3, the actions of the individuals to be evacuated in the neighborhood within a preset distance range can be collected by the sensor network in the area to be evacuated, so as to correct the influence parameters of the individuals in the neighborhood. Specifically, the correction can be made using a machine model, which will not be discussed in detail in this embodiment. For details, please refer to the methods in the prior art.

[0123] S103. Determine the current emotional level of each individual to be evacuated from the environmental impact parameters and the neighboring individual impact parameters using the first emotion-driven model.

[0124] The higher the current emotional level, the more panicked the individuals to be evacuated are; conversely, the lower the emotional level, the less panicked they are.

[0125] Specifically, the first emotion-driven model is as follows:

[0126] ;

[0127] in, For the current moment, For a historic moment, Based on the current emotional level, For historical sentiment levels, For the influence parameters of neighboring individuals, For environmental impact parameters, The social transmission coefficient, The environmental stimulus coefficient, The emotional self-regulation coefficient. , and This was deduced from experience. Used to quantify the impact on nearby individuals awaiting evacuation; the longer the contact time between these individuals and the evacuees, the greater the impact. The higher the value; Used to quantify the perceived level of danger; the stronger the perception of danger by the individual to be evacuated. The larger.

[0128] In one embodiment, evacuation guidance is provided in the evacuation area, such as broadcast guidance, signage guidance, indicator light guidance, and mobile terminal guidance. This serves two purposes: providing evacuation instructions to individuals and offering emotional reassurance. To further evaluate the impact of the evacuation guidance in the evacuation area on the emotional level of individuals, the scenario data also includes environmental guidance data. For example, environmental guidance data includes broadcast control variables. It is used to play soothing voice messages; environmental guidance data also includes visual guidance control quantities. This is used in conjunction with intelligent indicator lights to reduce the herd mentality of individuals awaiting evacuation; environmental guidance data also includes terminal push control quantities. This method is used for evacuation guidance, pushing planned routes to individuals to be evacuated via mobile devices. Other methods include:

[0129] The parameters of the reassurance effect of environmental guidance data on each individual to be evacuated are determined based on the preset relationship.

[0130] In practice, experience suggests a pre-defined relationship between environmental guidance data and soothing influence parameters. Generally, the larger the environmental guidance data, the larger the soothing influence parameter, and vice versa. For example, broadcast control parameters... The appeasement effect parameters can be adjusted. Impact on mood levels, individuals awaiting evacuation in high-risk emotional areas Increased to 0.7-0.9.

[0131] Based on this, step S103 further includes: determining the current emotional level of each individual to be evacuated from environmental influence parameters, neighborhood individual influence parameters, and reassurance influence parameters using a first emotion-driven model.

[0132] Specifically, the first emotion-driven model is as follows:

[0133] ;

[0134] in, For the current moment, For a historic moment, For individuals awaiting evacuation Current emotional level For individuals awaiting evacuation Historical sentiment levels For the influence parameters of neighboring individuals, For environmental impact parameters, Soothing the impact parameters, The social transmission coefficient, The environmental stimulus coefficient, To intervene and appease the coefficient, The emotional self-regulation coefficient. , , and All are constants. Based on empirical inference, The factors used to quantify information intervention are generally positively correlated with the effectiveness of broadcasting and the calming effect of crowd control personnel.

[0135] Regarding the first emotion-driven model in this embodiment, its derivation process is explained in detail below:

[0136] The first emotion-driven model treats each individual to be evacuated within a confined space as an independent "emotion-perceiving agent." The emotional evolution of each individual to be evacuated is influenced by three factors: their own psychological traits, the emotional contagion of neighboring individuals to be evacuated, and environmental stimuli. Based on this, the mathematical expression is constructed as follows:

[0137] ;

[0138] The definitions and physical meanings of each parameter are shown in the table below:

[0139]

[0140] S104. Determine the behavioral pattern of each individual to be evacuated based on their current emotional level.

[0141] Specifically, if the current emotional level is greater than or equal to the first preset level, the corresponding individual to be evacuated is identified as exhibiting irrational behavior; or, if the current emotional level is greater than the first preset level, the corresponding individual to be evacuated is identified as exhibiting rational behavior.

[0142] First preset level Characterizing the inherent psychological threshold of each individual to be evacuated. The value ranges from 0.3 to 0.7, but is not limited to this range; specific values ​​may vary from person to person and are randomly generated by the model. Specifically, when... When individuals awaiting evacuation enter a pattern of irrational behavior, At the same time, individuals awaiting evacuation maintained a rational behavior pattern. The differences in behavioral characteristics between the two patterns are shown in the table below:

[0143]

[0144] In some alternative implementations, emotional levels and behavioral patterns can be further subdivided:

[0145] behavioral patterns (Calm / Normal): Corresponding (Rational behavior pattern);

[0146] behavioral patterns (Anxiety / Mild Panic): Corresponding (Transitional behavior mode, speed slightly increased, still refer to safety tips);

[0147] behavioral patterns (High level of panic): Corresponding (Irrational behavior pattern, herd reinforcement);

[0148] behavioral patterns (Out of control / Extreme panic): Corresponding (Completely irrational behavior pattern, prone to pushing and shoving, and going against the flow).

[0149] It should be noted that the values ​​above are only examples and should be adjusted according to actual needs.

[0150] S105. Based on behavioral patterns, establish a comprehensive cost function for each individual to be evacuated, and combine it with a path planning algorithm to search for an evacuation path for each individual to be evacuated from the spatial model.

[0151] The path planning algorithm can be any of the existing technologies such as Dkjkstra's algorithm (a classic single-source shortest path algorithm), A* algorithm (an improved version of Dkjkstra's algorithm), or Floyd's algorithm (a dynamic programming algorithm), and the specific choice should be made according to the actual situation.

[0152] Specifically, the comprehensive cost function includes:

[0153] ;

[0154] in, For each individual to be evacuated, proceed to the exit. The shortest path distance, To reach the exit The degree of congestion of the path, The path hazard level, For path recommendation score, For following coefficient, To choose an export The proportion of individuals awaiting evacuation out of all individuals awaiting evacuation. Distance weights For crowding weight, Risk weight, To guide weights, To follow the weight, , , , All are constants.

[0155] In addition, real-time crowd distribution and movement speed are obtained through infrared counters and cameras, and online corrections are made. and Parameters, Environmental impact parameters Positive correlation Output based on the path planning algorithm. Output based on the path planning algorithm. It is positively correlated with mood level.

[0156] Weights of each item in the comprehensive cost function , , and Typically, corresponding to the behavior pattern, when the behavior pattern of the individuals to be evacuated is rational, the distance weight in the corresponding comprehensive cost function is reduced, and the crowding weight and guidance weight in the corresponding comprehensive cost function are increased; or, when the behavior pattern of the individuals to be evacuated is irrational, the distance weight in the corresponding comprehensive cost function is increased, and the crowding weight and guidance weight in the corresponding comprehensive cost function are reduced.

[0157] For example, under the rational behavior model, , , This represents a path planning algorithm that prioritizes the shortest path; under irrational behavior patterns: =0.6, =0.1, =0.1 indicates that the path planning algorithm prioritizes following others and weakens the congestion penalty. The above values ​​are only examples and are not intended to limit this embodiment.

[0158] In conducting crowd evacuation simulations, the inventors discovered that existing crowd evacuation simulations often employ microscopic motion models such as cellular automata and traditional social force models. These models simulate the behavior of individuals to be evacuated based solely on physical motion rules (such as distance, speed, and collision avoidance), neglecting the propagation characteristics of emotions and their driving effect on behavior. For example, traditional models assume that individuals to be evacuated always aim to "rationally choose the shortest path," failing to accurately reflect the actual situation in an emergency where panic leads to path confusion and sudden increases or decreases in speed. This results in significant deviations between simulation results and actual conditions, limiting their reference value. This embodiment bidirectionally couples emotional levels with behavioral patterns, using emotional levels as the core driving factor. It combines individual psychological traits (such as psychological thresholds and emotional regulation abilities), interpersonal interaction intensity (such as distance and contact frequency), and environmental stimuli (such as alarm sounds, smoke concentration, and congestion) to quantify the emotional levels of individuals to be evacuated in real time. These emotional levels are then directly mapped to behavioral patterns (such as speed changes, path selection preferences, and avoidance strategy adjustments), achieving dynamic linkage between emotions and behavior. For example, when the emotional level of an individual to be evacuated is below the psychological threshold, the shortest path is prioritized and a safe distance is maintained; when the emotional level of an individual to be evacuated is above the psychological threshold, the speed increases sharply (30%-80% higher than the rational state), herd behavior is enhanced (following the surrounding crowd), and the distance to avoid is reduced (physical contact is more likely to occur), realistically simulating the behavior pattern driven by emotions.

[0159] Furthermore, existing crowd evacuation simulations only focus on the physical congestion risk caused by excessive numbers of individuals to be evacuated in a sub-region, rarely predicting or warning of secondary risks arising from the spread of emotions, such as localized emotional outbursts, conflicts at exits, and irrational reversals. When the population density in a sub-region is below the threshold, but pushing and shoving have already occurred due to panic, existing systems cannot identify such potential risks. However, the path planning algorithm provided in this embodiment dynamically adjusts evacuation paths based on emotional levels and the congestion status of sub-regions. For example:

[0160] For individuals awaiting evacuation with low emotional levels (rational behavior), the shortest path is prioritized to ensure evacuation efficiency. For individuals awaiting evacuation with high emotional levels (irrational behavior), paths away from high-risk sub-areas are prioritized to avoid them being influenced by other individuals with high emotional levels. At the same time, they are guided to move to alternative exits with lower population density to balance exit pressure. When an exit becomes congested due to panic gathering, the system pushes alternative exit guidance information in real time (such as indicator light switching and broadcast prompts) to avoid excessive congestion at a single exit.

[0161] In one embodiment, to provide early warning of physical risk types in space, the behavioral data also includes the physical crushing force of each individual to be evacuated. This physical crushing force can be collected by sensors or simulated based on a spatial model. Based on this method, the method further includes:

[0162] When the physical pressure on the individual to be evacuated is greater than or equal to the first preset pressure and less than the second preset pressure, the current emotional level of the neighboring individuals to be evacuated within the preset distance range of the individual to be evacuated is increased; when the physical pressure on the individual to be evacuated is greater than or equal to the second preset pressure, an extreme risk warning is triggered, and the individual to be evacuated is transformed into an obstacle node in the spatial model in order to replan the evacuation path of the neighboring individuals to be evacuated within the preset distance range of the individual to be evacuated.

[0163] Specifically, based on the magnitude of the physical compressive force experienced by the individuals to be evacuated, their states are categorized into normal state, injured state, and fatal state, with the following conversion rules:

[0164] The first preset pressure is set to 800 N / m, and the second preset pressure is set to 1600 N / m. When the physical pressure on the individual to be evacuated is less than 800 N / m, the individual is in a normal state, can move normally, and exhibits normal emotional development. When the physical pressure on the individual to be evacuated is greater than or equal to 800 N / m but less than or equal to < 1600 N / m, the individual is in an injured state. Due to increased pain and fear, the individual's movement speed decreases by 50%, and their emotional level increases by 0.2. Simultaneously, the social contagion coefficient to neighboring individuals to be evacuated also increases. A temporary 20% increase is applied. When the physical compressive force on an individual to be evacuated exceeds 1600 N / m, that individual is considered injured or killed. Upon ceasing movement, the individual transforms into an obstacle node in the spatial model, and the system immediately triggers an extreme risk warning. Simultaneously, the evacuation paths of neighboring individuals within a preset distance range of the individual to be evacuated are replanned. It should be noted that the above values ​​are for illustrative purposes only; specific configurations should be made according to actual needs.

[0165] During the crowd evacuation simulation process, the inventors discovered that existing crowd evacuation simulation methods have high computational complexity and do not incorporate the dynamic evolutionary characteristics of emotion transmission, making it difficult to capture sudden risk changes caused by the spread of emotions in real time, such as congestion in a sub-area due to panic within a short period of time. If the early warning lags behind the actual situation on site (e.g., a lag of more than 10 seconds), it cannot provide effective decision support for management personnel. In this embodiment, by constructing a multi-dimensional risk early warning mechanism based on emotional level and physical pressure, simultaneous identification of emotional and physical risks is achieved. Furthermore, this embodiment can also construct a multi-level early warning mechanism:

[0166] 1. Emotional Risk Warning: When the average emotional level of a sub-area exceeds the first risk threshold (e.g., 0.6, with a value range of [0,1]) or the diffusion rate of the average emotional level exceeds the second risk threshold (e.g., an increase of 0.2 per minute), the sub-area is determined to be a high-emotional-risk sub-area. The system triggers an emotional risk warning (e.g., audio-visual prompts, broadcast reassurance instructions), prompting the counselors to go to the sub-area to guide the crowd's emotions and prevent the escalation of irrational behavior.

[0167] 2. Physical Risk Warning: When the density of individuals to be evacuated in a certain sub-area exceeds the third risk threshold (e.g., 5 people / ㎡, adjusted according to space type), or the local compression force reaches the fourth risk threshold (e.g., 800 N / m, lower than the casualty threshold of 1600 N / m), it is determined to be a high physical risk sub-area. The system triggers a physical risk warning to dynamically adjust the evacuation route and guide some individuals to be evacuated to the backup exit.

[0168] 3. Extreme Risk Warning: When casualties occur in a certain area (compression force reaches 1600 N / m) or panic spreads throughout the area (average mood level exceeds 0.8), an extreme risk warning is triggered. The system immediately notifies management personnel to block entrances, increase evacuation forces, and activate emergency broadcast to issue unified evacuation instructions to prevent the risk from spreading.

[0169] In one embodiment, considering the emotional exhaustion of individuals awaiting evacuation due to prolonged periods of high emotional levels, an emotional fatigue mechanism is also established for them. This mechanism is set based on the duration of the high emotional level experienced by each individual. Specifically, the behavioral data includes the historical emotional level of each individual. The method further includes:

[0170] If the duration of the individual's historical emotional level being greater than or equal to the first preset level is greater than or equal to the preset duration, the current emotional level of the individual to be evacuated is determined by the second emotion-driven model.

[0171] Specifically, the second emotion-driven model obtains the current emotion level by attenuating historical emotion levels using preset attenuation parameters and preset durations, expressed as:

[0172] ;

[0173] in, Individuals to be evacuated Historical emotional level is greater than or equal to the first preset level Duration For preset duration, This is the preset attenuation parameter.

[0174] When the duration of a person's elevated emotional state during evacuation is greater than or equal to a preset duration, their emotion regulation ability is enhanced, and the rate of emotional decline increases. For example, if a person's elevated emotional state lasts for 40 seconds, and the preset duration is set to... If the emotional level decreases by a certain number of seconds, it will decrease further on top of the existing decline. The study simulated the panic relief trend of individuals awaiting evacuation due to emotional exhaustion.

[0175] In one embodiment, the behavioral data of the individuals to be evacuated includes spatial location and motion state as described in the above embodiments. In addition to data collected by sensors, this data is also updated based on microscopic motion model simulations. (See [link to relevant documentation]). Figure 4 After step S105, the method further includes:

[0176] S106. Determine the motion state of each individual to be evacuated based on the micro motion model, and update the spatial position of each individual to be evacuated based on the motion state.

[0177] Specifically, the microscopic motion model can be any of the following: cellular automata model or social force model.

[0178] In an optional implementation, step S106, which determines the motion state of each individual to be evacuated based on a cellular automata model, requires setting motion rules corresponding to behavioral characteristics of emotional levels to ensure consistency between the outputs of the two models.

[0179]

[0180] In an optional implementation, step S106, which determines the motion state of each individual to be evacuated based on a social force model, includes:

[0181] The current emotional level of each individual to be evacuated is mapped to multiple moderating factors in the social force model to determine the movement state of each individual to be evacuated.

[0182] Among them, the regulatory factors include self-motivation, interpersonal interaction, boundary forces, and emotional drive.

[0183] Specifically, the social force model provided in this embodiment introduces an additional force driven by emotions on the basis of the traditional social force model. The total force on the individual to be evacuated consists of multiple moderating factors, including self-motivation force, interpersonal interaction force, boundary force, and additional force driven by emotions. The mathematical expression is as follows:

[0184] ;

[0185] in, For individuals awaiting evacuation Total force, For individuals awaiting evacuation Self-driving force For individuals awaiting evacuation Interpersonal skills For individuals awaiting evacuation Boundary forces, For individuals awaiting evacuation The emotional driving force. The calculation logic and physical meaning of each moderating factor are as follows:

[0186] 1. Self-driving force

[0187] This reflects the willingness of individuals awaiting evacuation to move towards the target exit, and is positively correlated with emotional level; that is, the higher the emotional level, the stronger the drive of the individuals awaiting evacuation towards the exit. The expression is:

[0188] ;

[0189] in, For individuals awaiting evacuation The mass (valued at 50-80 kg to simulate individual weight differences); For individuals awaiting evacuation The expected speed changes dynamically with emotional level. ( The expected velocity under rational conditions, such as 1.2 m / s. (This is the panic speed gain coefficient, with a value of 0.3-0.8). For individuals awaiting evacuation At any moment The actual speed; Adjust reaction time to speed (values ​​range from 0.1 to 1 second; the higher the emotional level, the faster the reaction). The smaller the value, the faster the speed adjustment, usually in line with the previous value. (To maintain consistency), the values ​​above are for illustrative purposes only and should be adjusted according to the actual situation.

[0190] 2. Interpersonal skills

[0191] This reflects the interaction between the individual to be evacuated, k, and their neighboring individuals, j, including psychological repulsion (the force maintaining distance when not in contact), physical pressure (the interaction force upon contact), and friction (relative sliding resistance upon contact). Higher emotional levels result in weaker psychological repulsion and higher tolerance for pressure. The expression is:

[0192] ;

[0193] in, The psychological repulsion coefficient decreases with increasing emotional level. ( (This refers to the psychological repulsion coefficient under rational conditions, such as 200 N). The psychological repulsive force attenuation coefficient (valued between 0.1 and 0.3 meters, reflecting the rate attenuation of repulsive force with distance). , Let K be the body radius of the individual to be evacuated (k) and the neighboring individual to be evacuated (j) (values ​​range from 0.25 to 0.35 meters, simulating the difference in shoulder width). Let k be the center distance between the individual to be evacuated and the neighboring individual j to be evacuated; The normal vector pointing from the individual to be evacuated k to the neighboring individual to be evacuated j (perpendicular to the line connecting the centers of the individual to be evacuated k and the neighboring individual to be evacuated j, reflecting the direction of the repulsive force); The compression coefficient (valued at 5000-8000 N / m, reflecting the force transmission efficiency during physical contact); Let k be the velocity of the individual to be evacuated relative to its neighboring individual j. The tangent vector (parallel to the direction perpendicular to the line connecting the centers of the individual to be evacuated k and the neighboring individual to be evacuated j, reflecting the direction of friction) is shown above as an example only, and should be adjusted according to the actual situation.

[0194] 3. Boundary forces

[0195] This reflects the interaction between the individual to be evacuated (k) and fixed boundaries such as walls and obstacles. The calculation logic is similar to that of interpersonal interaction forces, but the parameters are adjusted according to the type of boundary (such as the psychological repulsion coefficient of the wall). (The repulsive force coefficient between individuals k to be evacuated is 1.5 times the coefficient of repulsion between individuals k to be evacuated, ensuring a safe distance between individuals k to be evacuated and the wall). The expression is:

[0196] ;

[0197] in, Let k be the distance from the boundary to the individual to be evacuated. Let be the normal vector pointing from the boundary to the individual k to be evacuated; , , The parameters are related to the boundary (preset according to the boundary material and type, such as the different parameters for glass walls and solid walls). The above values ​​are only examples and should be adjusted according to the actual situation.

[0198] 4. Emotion-driven additional force

[0199] Reflecting the irrational movement trends triggered by panic, individuals (k) with high emotional levels are more likely to follow the surrounding crowd, exacerbating herd behavior. The expression is:

[0200] ;

[0201] in, The emotional drive coefficient (valued at 10-30 N·s / m, reflecting the intensity of panic-driven follower behavior). The unit vector is the direction of the target exit (to ensure that the following behavior is in the direction of evacuation and avoids going against the flow). The above values ​​are only examples and should be adjusted according to the actual situation.

[0202] Furthermore, to ensure that path planning and following behavior align with social impact (i.e., to resonate with the emotionally driven additional force), the following coefficient in the following comprehensive function... Direct mapping to Emotional drive coefficient ,Right now ( (Using a baseline coefficient) to ensure consistency between following behavior and emotion-driven logic, and to avoid model conflicts.

[0203] S107. Determine whether each individual to be evacuated has left the evacuation area based on the updated spatial location. If the determination is no, continue to execute step S101 until each individual to be evacuated leaves the evacuation area.

[0204] In one embodiment, in order to respond quickly to the crowd evacuation simulation method provided in this disclosure, this embodiment combines different evacuation scenarios to perform multi-time-dimensional simulations, specifically:

[0205] Pre-emptive rehearsals: Based on historical scenario data, simulate evacuation scenarios under different ignition points, pedestrian flow, and entrance / exit opening / closing strategies to build a contingency plan library.

[0206] Real-time adaptation: When an accident occurs, the system matches the most similar scenario in the contingency plan database with real-time scenario data and provides guidance in conjunction with evacuation route planning.

[0207] Post-event review: Comparing simulation results with actual conditions, adjustments were made to the parameters of various models, including the emotion-driven model, the comprehensive cost function, and the path planning algorithm. Specifically, Model Predictive Control (MPC) was employed to adjust these parameters based on simulation results and actual conditions, ensuring the optimization objectives were achieved. The control cycle and time step of MPC were consistent. The optimization objectives of MPC included shortening the total evacuation time, reducing local peak density, decreasing the proportion of highly panicked individuals, and improving the balance of exit utilization.

[0208] See the diagram below. Figure 5 The crowd evacuation simulation method provided in this embodiment will be described in detail through a specific simulation example:

[0209] Step S1: Initialization. Specifically:

[0210] 1. Import the building floor plan of the closed area, and construct a "grid + graph" spatial model through LiDAR scanning and camera image recognition (adaptive modeling of complex areas).

[0211] 2. Establish a spatial location model for hazardous materials and a hazardous area model;

[0212] 3. Initialize the behavioral data of the individuals to be evacuated (consistent with the Agent attribute, initial emotional level). );

[0213] 4. Configure environmental guidance data such as broadcasting and guidance systems (parameters to mitigate impact). ).

[0214] Step S2: Environmental Hazard Update. Specifically:

[0215] 1. Based on the hazardous field model, update the hazard index of each grid by combining the concentration of hazardous materials collected by various sensors such as smoke and temperature. ;

[0216] 2. Calculate the accessibility of exits and passageways (grid hazard index of exits or passageways). (≥0.8, deemed impassable).

[0217] 3. Update environmental impact parameters synchronously. .

[0218] Step S3: Information and guidance strategy update. Specifically:

[0219] 1. Based on the risk identification results, update the broadcast guidance content (play reassuring voice messages in high emotional risk areas and path guidance voice messages in high physical risk areas), and obtain the broadcast control quantity. ;

[0220] 2. Adjust the direction of the signage and the status of the indicator lights to obtain visual guidance control parameters. ;

[0221] 3. Obtain terminal push control volume by guiding push path suggestions through mobile terminals. .

[0222] Step S4: Individual danger perception and emotional update. Specifically:

[0223] 1. Calculate the individuals to be evacuated. of and ;

[0224] 2. Substitute into the emotion-driven model to calculate the emotion level. ;

[0225] 3. Determine behavioral patterns based on emotional levels.

[0226] Step S5: Exit and Route Selection. Specifically:

[0227] 1. Adjust the overall cost function based on the current emotional level of the individuals to be evacuated. ;

[0228] 2. The A* algorithm is used to calculate evacuation routes. Rational behavior patterns prioritize the shortest path, while irrational behavior patterns prioritize paths that are away from high-emotional-risk areas.

[0229] 3. Output the evacuation routes for the individuals to be evacuated.

[0230] Step S6: Update motion status. Specifically:

[0231] 1. Choose either a social force model or a cellular automata model to calculate the motion state based on individual behavioral patterns. The social force model uses the total force formula to calculate the motion state and spatial location, while the cellular automata model updates the motion state and spatial location of the individuals to be evacuated based on the motion rules corresponding to the behavioral characteristics of emotional levels.

[0232] 2. Handle conflicts (multiple individuals vying for the same grid) and collisions (compression force ≥ 800 N / m), and update the status of the individuals to be evacuated;

[0233] 3. Update the movement status and spatial location of all individuals to be evacuated.

[0234] Step S7: Statistics and Evaluation. Specifically:

[0235] 1. Key statistical indicators: number of people not evacuated, number of people evacuated, average evacuation time, maximum evacuation time;

[0236] 2. Assess risk status: Assess each sub-area based on factors such as the emotional state of individuals to be evacuated, population density distribution, local pressure, and exit utilization rate.

[0237] 3. Compare with preset thresholds: Issue warnings based on the status of individuals to be evacuated, and provide environmental guidance based on the evacuation route.

[0238] Step S8: Iterate through the loop. Specifically:

[0239] Repeat steps S2-S7 until all personnel have been evacuated or an extreme risk warning is triggered (simulation termination condition).

[0240] 2. Hazardous area calibration: Real-time updates using data from smoke, temperature, and hazardous gas sensors. Synchronous correction (Intensity of environmental stimuli);

[0241] 3. Mood parameter correction: Using heart rate data collected by wearable devices to assist in correction. (Emotional level) (Emotional self-regulation coefficient) (Intervention and reassurance coefficient).

[0242] An exemplary embodiment of this disclosure provides a crowd evacuation simulation device, see [link to example]. Figure 6 The device includes:

[0243] The acquisition module 61 is used to acquire scene data in the area to be evacuated; the scene data includes sensor data of hazardous materials and behavioral data of each individual to be evacuated in the crowd;

[0244] The mapping module 62 is used to map scene data to the spatial model of the area to be evacuated, so as to determine the environmental impact parameters and the impact parameters of neighboring individuals for each individual to be evacuated.

[0245] The first level determination module 63 is used to determine the current emotional level of each individual to be evacuated from environmental influence parameters and neighborhood individual influence parameters through a first emotion-driven model.

[0246] The pattern determination module 64 is used to determine the behavioral pattern of each individual to be evacuated based on the current emotional level.

[0247] Planning module 65 is used to establish a comprehensive cost function for each individual to be evacuated based on behavioral patterns, and to search for evacuation paths for each individual to be evacuated from the spatial model in conjunction with a path planning algorithm.

[0248] In one embodiment, the scene data further includes map data of the area to be evacuated, and the device further includes:

[0249] The model building module is used to build a spatial model based on map data. The spatial model includes at least a grid set, a node set, and an edge set. Each grid cell in the grid set represents a passable or impassable sub-region in the area to be evacuated. Each node in the node set represents a preset sub-region in the area to be evacuated. Each edge in the edge set represents a passageway between two preset sub-regions.

[0250] In one embodiment, the sensing data includes the concentration of hazardous materials, and the mapping module 62 is further used for:

[0251] Hazardous materials and each individual to be evacuated are mapped into a spatial model to determine their spatial locations;

[0252] Based on the hazardous field model, the concentration of hazardous materials and the spatial location of hazardous materials and each individual to be evacuated are calculated to obtain the environmental impact parameters for each individual to be evacuated.

[0253] In one embodiment, the behavioral data includes the historical emotional level of each individual to be evacuated, and the mapping module 62 is further used for:

[0254] Each individual to be evacuated is mapped into a spatial model to determine the spatial location of each individual.

[0255] Identify the neighboring individuals to be evacuated within a preset distance range for each individual to be evacuated;

[0256] Based on the distance between the neighboring individuals to be evacuated and each other, the emotional level of the neighboring individuals to be evacuated is integrated and averaged to obtain the influence parameter of the neighboring individuals for each individual to be evacuated.

[0257] In one embodiment, the scene data further includes environmental guidance data; the device also includes:

[0258] The guidance module is used to: determine the reassurance impact parameters of environmental guidance data on each individual to be evacuated based on preset relationships;

[0259] The first level determination module 63 is also used for:

[0260] The current emotional level of each individual to be evacuated is determined using the first emotion-driven model, based on environmental impact parameters, neighborhood individual impact parameters, and reassurance impact parameters.

[0261] In one embodiment, the first emotion-driven model includes:

[0262] ;

[0263] Where t represents a historical moment. For the current moment, Let be the current emotional level of the k-th individual to be evacuated at the current moment. For the influence parameters of neighboring individuals, For environmental impact parameters, To appease the affected parameters, The emotional contagion coefficient, The environmental stimulus coefficient, To intervene and appease the coefficient, The emotional self-regulation coefficient. , and All are constants.

[0264] In one embodiment, the behavioral data also includes the physical compressive force of each individual to be evacuated, and the device further includes:

[0265] The pressure adjustment module is used to raise the current emotional level of neighboring individuals within a preset distance range of the individual to be evacuated when the physical pressure of the individual to be evacuated is greater than or equal to the first preset pressure and less than the second preset pressure; or to trigger an extreme risk warning when the physical pressure of the individual to be evacuated is greater than or equal to the second preset pressure, and to transform the individual to be evacuated into an obstacle node in the spatial model so as to replan the evacuation path of neighboring individuals within a preset distance range of the individual to be evacuated.

[0266] In one embodiment, the behavioral data also includes the historical emotional level of each individual to be evacuated, and the device further includes:

[0267] The second level determination module is used to determine the current emotional level of the individual to be evacuated by means of a second emotion-driven model when the duration of the historical emotional level of the individual to be evacuated is greater than or equal to the first preset level and is greater than or equal to the preset duration. The second emotion-driven model obtains the current emotional level by attenuating the historical emotional level by using preset attenuation parameters and preset duration.

[0268] In one embodiment, the pattern determination module 64 is further configured to:

[0269] When the current emotional level is greater than or equal to the first preset level, the corresponding individual to be evacuated is identified as exhibiting an irrational behavior pattern.

[0270] Alternatively, if the current emotional level is greater than the first preset level, the corresponding individual to be evacuated is identified as exhibiting a rational behavior pattern.

[0271] In one embodiment, the comprehensive cost function includes:

[0272] ;

[0273] in, For each individual to be evacuated, proceed to the exit. The shortest path distance, To reach the exit The degree of congestion of the path, The path hazard level, For path recommendation score, For following coefficient, To choose an export The proportion of individuals awaiting evacuation out of all individuals awaiting evacuation. Distance weights For crowding weight, Risk weight, To guide weights, To follow the weight, , , , All are constants.

[0274] In one embodiment, the planning module 65 is further configured to:

[0275] When the behavior pattern of the individuals to be evacuated is rational behavior, the distance weight in the corresponding comprehensive cost function is reduced, and the crowding weight and guidance weight in the corresponding comprehensive cost function are increased.

[0276] Alternatively, when the behavior pattern of the individuals to be evacuated is irrational, increase the distance weight in the corresponding comprehensive cost function and decrease the crowding weight and guidance weight in the corresponding comprehensive cost function.

[0277] In one embodiment, the apparatus further includes:

[0278] The position and status update module is used to determine the motion state of each individual to be evacuated based on the micro motion model, update the spatial position of each individual to be evacuated based on the motion state, and determine whether each individual to be evacuated has left the evacuation area based on the updated spatial position. If the determination is negative, the acquisition module 61 continues to be executed until each individual to be evacuated leaves the evacuation area.

[0279] In one embodiment, the micro-motion model includes a social force model;

[0280] The location and status update module is also used to map the current emotional level of each individual to be evacuated to multiple moderating factors in the social force model to determine the movement status of each individual to be evacuated; the moderating factors include self-driving force, interpersonal interaction force, boundary force, and emotional driving additional force.

[0281] In one embodiment, the emotion-driven additional force includes:

[0282] ;

[0283] in, For individuals awaiting evacuation The added force of emotion-driven action Emotion-driven weights Let be the unit vector in the export direction. For individuals awaiting evacuation speed, For individuals awaiting evacuation speed, For the preset distance range, For individuals awaiting evacuation with individuals to be evacuated The distance between them.

[0284] An exemplary embodiment of this disclosure provides a crowd evacuation simulation system, see [link to example]. Figure 7 The system includes a data acquisition layer 71, a data processing layer 72, and an evacuation layer 73.

[0285] The data acquisition layer 71 and the evacuation layer 73 are both connected to the data processing layer. The data acquisition layer 71 is used to collect scene data. The data processing layer 72 is loaded with a crowd evacuation simulation method as described in any of the first aspects. The data processing layer 72 is used to determine the evacuation path based on the scene data. The evacuation layer 73 is used to evacuate the crowd based on the evacuation path.

[0286] In this embodiment, a three-layer architecture of data acquisition layer - data processing layer - evacuation layer is adopted to achieve deep integration of model and hardware:

[0287] Specifically, the data acquisition layer 71 includes individual status acquisition devices (e.g., infrared counters, high-definition cameras, wearable devices), environmental status acquisition devices (e.g., smoke sensors, sound sensors, lidar), and a data transmission module (e.g., 5G / Wk-Fk 6, latency ≤ 1 second). The data acquisition layer 71 provides input to the data processing layer 72.

[0288] Specifically, the data processing layer 72 includes edge computing nodes (for preprocessing scene data), a cloud computing platform (for distributed computing and data storage), an algorithm module (for simulating core models such as emotion propagation, motion state simulation, path planning, and risk warning), and an extension module. The data processing layer is used to determine evacuation routes and issue warnings based on scene data.

[0289] Specifically, evacuation layer 73 includes management personnel terminals (such as monitoring screens and mobile apps), public early warning equipment (audio and visual alarms, emergency broadcasts, and smart indicator lights), and evacuation guidance modules (navigation screens and mobile terminal push notifications), which convert model outputs into early warning information and evacuation instructions.

[0290] The following is a specific equipment configuration for a crowd evacuation simulation system, which can be adjusted according to specific needs:

[0291]

[0292] The crowd evacuation simulation system provided in this embodiment has the following advantages:

[0293] 1. The simulation realism is significantly improved. Through the coupling and unified parameter system of multiple models such as emotion-driven model, path planning algorithm and micro motion model, the simulation can realistically simulate the chain reaction of "emotion spread → behavioral alienation → increased congestion" in the evacuation area. This solves the defect of traditional models that only calculate physical phenomena and not emotions. The deviation rate between the simulation results and the actual scene is reduced by more than 35%.

[0294] 2. Enhanced real-time response capabilities: By combining edge computing and cloud-based collaborative architecture, the delay in updating sentiment levels and issuing risk warnings is reduced, enabling rapid detection of risk mutations caused by the spread of sentiment.

[0295] 3. Dual optimization of evacuation efficiency and safety: Based on emotion-driven path planning, the overall evacuation time is shortened, and potential risks such as emotional outbursts and physical congestion can be identified in advance, effectively avoiding stampedes.

[0296] 4. It has strong scene adaptability, supports the establishment of spatial models for complex and irregular areas, and is suitable for various enclosed spaces such as shopping malls and subways. The hardware can reuse existing security systems, reducing deployment costs and facilitating large-scale promotion.

[0297] 5. Full-process management support: Through a closed-loop mechanism of pre-event rehearsal, in-event early warning, and post-event review, it provides full-cycle support for the safety management of enclosed spaces. It is applicable to both daily contingency plan optimization and emergency response.

[0298] Figure 8 This is a structural diagram of an electronic device provided in an example embodiment of the present disclosure. Figure 8 The electronic device 800 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0299] like Figure 8 As shown, the electronic device 800 can be manifested in the form of a general-purpose computing device, such as a server device. The components of the electronic device 800 may include, but are not limited to: at least one processor 801, at least one memory 802, and a bus 803 connecting different system components (including memory 802 and processor 801).

[0300] The 803 bus includes a data bus, an address bus, and a control bus.

[0301] The memory 802 may include volatile memory, such as random access memory (RAM) 8021 and / or cache memory 8022, and may further include read-only memory (ROM) 8023.

[0302] The memory 802 may also include a program tool 8025 (or utility) having a set (at least one) program module 8024, such program module 8024 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network scenario.

[0303] The processor 801 executes various functional applications and data processing, such as the methods provided in any of the above embodiments, by running computer programs stored in the memory 802.

[0304] Electronic device 800 can also communicate with one or more external devices 804. This communication can be made via input / output (K / O) interface 805. Furthermore, the model-generated electronic device 800 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 806. As shown in the figure, network adapter 806 communicates with other modules of the model-generated electronic device 800 via bus 803. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the model-generated electronic device 800, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAKD (remote array) systems, tape drives, and data backup storage systems.

[0305] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0306] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method provided in any of the above embodiments.

[0307] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0308] In possible implementations, embodiments of this disclosure can also be implemented as a program product including program code, which, when the program product is run on a terminal device, causes the terminal device to execute the method implementing any of the above embodiments.

[0309] The program code for executing this disclosure can be written in any combination of one or more programming languages. The program code can be executed entirely on a user device, partially on a user device, as a standalone software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0310] While specific embodiments of this disclosure have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of this disclosure is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of this disclosure, but all such changes and modifications shall fall within the scope of protection of this disclosure.

Claims

1. A method for simulating crowd evacuation, characterized in that, The method includes: Acquire scene data in the area to be evacuated; the scene data includes sensor data of hazardous materials and behavioral data of each individual to be evacuated in the crowd; The scene data is mapped to the spatial model of the area to be evacuated in order to determine the environmental impact parameters and the impact parameters of neighboring individuals for each individual to be evacuated. The current emotional level of each individual to be evacuated is determined using a first emotion-driven model based on environmental impact parameters and neighborhood individual impact parameters. Determine the behavioral pattern of each individual to be evacuated based on the current emotional level; Based on the behavioral patterns, a comprehensive cost function is established for each individual to be evacuated, and a path planning algorithm is used to search for an evacuation path for each individual to be evacuated from the spatial model.

2. The crowd evacuation simulation method as described in claim 1, characterized in that, The scene data also includes map data of the area to be evacuated; The method further includes: A spatial model is established based on the map data; the spatial model includes at least a grid set, a node set, and an edge set. Each grid cell in the grid set represents a passable or impassable sub-region in the area to be evacuated. Each node in the node set represents a preset sub-region in the area to be evacuated. Each edge in the edge set represents a passageway between two preset sub-regions.

3. The crowd evacuation simulation method as described in claim 1, characterized in that, The sensor data includes the concentration of hazardous materials; The step of mapping the scene data to the spatial model of the area to be evacuated, in order to determine the environmental impact parameters of each individual to be evacuated, includes: Map the hazardous materials and each individual to be evacuated to the spatial model to determine the spatial location of the hazardous materials and each individual to be evacuated; The concentration of the hazardous materials and the spatial location of the hazardous materials and each individual to be evacuated are calculated based on the hazardous field model to obtain the environmental impact parameters of each individual to be evacuated.

4. The crowd evacuation simulation method as described in claim 1, characterized in that, The behavioral data includes the historical emotional level of each of the individuals to be evacuated; The step of mapping the scene data to the spatial model of the area to be evacuated, in order to determine the influence parameters of neighboring individuals for each individual to be evacuated, includes: Each individual to be evacuated is mapped into the spatial model to determine the spatial location of each individual to be evacuated; Determine the neighboring individuals to be evacuated within a preset distance range for each individual to be evacuated; Based on the distance between the neighboring individuals to be evacuated and each of the individuals to be evacuated, the emotional levels of the neighboring individuals to be evacuated are integrated and averaged to obtain the influence parameters of the neighboring individuals for each individual to be evacuated.

5. The crowd evacuation simulation method as described in claim 1, characterized in that, The scenario data also includes environmental guidance data; The method further includes: The environmental guidance data is used to determine the reassurance effect parameters on each individual to be evacuated based on a preset relationship. The step of determining the current emotional level of each individual to be evacuated from environmental influence parameters and neighboring individual influence parameters using the first emotion-driven model includes: The first emotion-driven model determines the current emotional level of each individual to be evacuated from environmental influence parameters, neighborhood individual influence parameters, and reassurance influence parameters.

6. The crowd evacuation simulation method as described in claim 5, characterized in that, The first emotion-driven model includes: ； in, For the current moment, For a historic moment, For individuals awaiting evacuation, For individuals awaiting evacuation Current emotional level For individuals awaiting evacuation Historical sentiment levels For the influence parameters of neighboring individuals, For environmental impact parameters, Soothing the impact parameters, The social transmission coefficient, The environmental stimulus coefficient, To intervene and appease the coefficient, The emotional self-regulation coefficient. , , and All are constants.

7. The crowd evacuation simulation method as described in claim 1, characterized in that, The behavioral data also includes the physical compressive force of each of the individuals to be evacuated; The method further includes: When the physical pressure on the individual to be evacuated is greater than or equal to the first preset pressure and less than the second preset pressure, the current emotional level of the neighboring individuals to be evacuated within a preset distance range of the individual to be evacuated is increased. When the physical pressure on the individual to be evacuated is greater than or equal to the second preset pressure, an extreme risk warning is triggered, and the individual to be evacuated is transformed into an obstacle node in the spatial model, so as to replan the evacuation path of the neighboring individuals to be evacuated within a preset distance range of the individual to be evacuated.

8. The crowd evacuation simulation method as described in claim 1, characterized in that, The behavioral data also includes the historical emotional level of each of the individuals to be evacuated; The method further includes: If the duration of the historical emotional level of the individual to be evacuated being greater than or equal to the first preset level is greater than or equal to the preset duration, the current emotional level of the individual to be evacuated is determined by the second emotion-driven model; the second emotion-driven model obtains the current emotional level by attenuating the historical emotional level with preset attenuation parameters and preset duration.

9. The crowd evacuation simulation method as described in claim 1, characterized in that, Determining the behavioral pattern of each individual to be evacuated based on the current emotional level includes: When the current emotional level is greater than or equal to the first preset level, the corresponding individual to be evacuated is determined to be in an irrational behavior pattern. Alternatively, if the current emotional level is greater than a first preset level, the corresponding individual to be evacuated is determined to be in a rational behavior pattern.

10. The crowd evacuation simulation method as described in claim 1, characterized in that, The comprehensive cost function includes: ； in, For each of the individuals to be evacuated to the exit The shortest path distance, For access to the said exit The degree of congestion of the path, The path hazard level, For path recommendation score, For following coefficient, To select the exit The proportion of the individuals to be evacuated to all the individuals to be evacuated. Distance weights For crowding weight, Risk weight, To guide weights, To follow the weight, , , , All are constants.

11. The crowd evacuation simulation method as described in claim 10, characterized in that, The establishment of a comprehensive cost function for each individual to be evacuated based on the behavioral pattern includes: When the behavior pattern of the individuals to be evacuated is a rational behavior pattern, the distance weight in the corresponding comprehensive cost function is reduced, and the crowding weight and guidance weight in the corresponding comprehensive cost function are increased. Alternatively, when the behavior pattern of the individuals to be evacuated is an irrational behavior pattern, the distance weight in the corresponding comprehensive cost function is increased, and the crowding weight and guidance weight in the corresponding comprehensive cost function are decreased.

12. The crowd evacuation simulation method as described in claim 1, characterized in that, After establishing a comprehensive cost function for each individual to be evacuated based on the behavioral pattern, and searching for an evacuation path for each individual to be evacuated from the spatial model using a path planning algorithm, the process includes: The motion state of each individual to be evacuated is determined based on the micro-motion model, and the spatial position of each individual to be evacuated is updated based on the motion state. Determine whether each individual to be evacuated has left the evacuation area based on the updated spatial location; If the determination is negative, continue with the step of obtaining scene data in the area to be evacuated until each individual to be evacuated leaves the area to be evacuated.

13. The crowd evacuation simulation method as described in claim 12, characterized in that, The micro-motion model includes a social force model; Determining the motion state of each individual to be evacuated based on the micro-motion model includes: The current emotional level of each individual to be evacuated is mapped to multiple moderating factors in the social force model to determine the movement state of each individual to be evacuated; the moderating factors include self-driving force, interpersonal interaction force, boundary force, and emotional driving additional force.

14. The crowd evacuation simulation method as described in claim 13, characterized in that, The emotionally driven additional force includes: ； in, For individuals awaiting evacuation The added force of emotion-driven action Emotion-driven weights Let be the unit vector in the export direction. For individuals awaiting evacuation speed, For individuals awaiting evacuation speed, For the preset distance range, For individuals awaiting evacuation with individuals to be evacuated The distance between them.

15. A crowd evacuation simulation device, characterized in that, The device includes: The acquisition module is used to acquire scene data in the area to be evacuated; the scene data includes sensor data of hazardous materials and behavioral data of each individual to be evacuated in the crowd; The mapping module is used to map the scene data to the spatial model of the area to be evacuated, so as to determine the environmental impact parameters and the impact parameters of neighboring individuals for each individual to be evacuated. The first level determination module is used to determine the current emotional level of each individual to be evacuated from environmental influence parameters and neighborhood individual influence parameters using a first emotion-driven model. The pattern determination module is used to determine the behavioral pattern of each individual to be evacuated based on the current emotional level. The planning module is used to establish a comprehensive cost function for each individual to be evacuated based on the behavioral pattern, and to search for an evacuation path for each individual to be evacuated from the spatial model in combination with a path planning algorithm.

16. A crowd evacuation simulation system, characterized in that, The system includes a data acquisition layer, a data processing layer, and an evacuation layer; Both the data acquisition layer and the evacuation layer are connected to the data processing layer. The data acquisition layer is used to collect scene data. The data processing layer is loaded with the crowd evacuation simulation method as described in any one of claims 1-14. The data processing layer is used to determine the evacuation path based on the scene data. The evacuation layer is used to evacuate the crowd based on the evacuation path.

17. An electronic device comprising a memory, a processor, and a computer program stored in the memory and for running on the processor, characterized in that, When the processor executes the computer program, it implements the crowd evacuation simulation method as described in any one of claims 1-14.

18. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the crowd evacuation simulation method as described in any one of claims 1-14.

19. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the crowd evacuation simulation method as described in any one of claims 1-14.

Images