Rescue scenario simulation method, device and equipment based on behavior tree

Abstract

The present invention provides a rescue scenario simulation method, device and equipment based on a behavior tree, which is used to solve the technical problems of low simulation level of disaster rescue scenarios and poor disaster prevention capabilities in related technologies. The method includes: analyzing the rescue scenario to be simulated to obtain scene information of the rescue scenario; determining a calculation model of the scenario based on the scene information, and the calculation model is used to simulate the state of the rescue scenario according to the scene information; generating multiple disaster-causing factor agents and multiple rescue force agents based on the scene information and the calculation model; configuring behavior trees for the multiple disaster-causing factor agents and the multiple rescue force agents, and simulating the rescue scenario.

Description

Technical Field

[0001] The present invention relates to the field of simulation, and in particular to a rescue scenario simulation method, device and equipment based on behavior tree. Background Art

[0002] Forest fire is the most serious type of forest disaster. It is sudden, destructive, wide-ranging, difficult to control, and difficult to extinguish and rescue. Once it occurs, it will cause huge losses to forest resources and seriously affect people's lives and property safety.

[0003] A large proportion of forest fires are caused by untimely or inappropriate firefighting plans, which miss the optimal time to extinguish a small fire, leading to its development into a large-scale blaze. During firefighting, humans, the environment, and the fire form a system. Therefore, by simulating human behavior during a forest fire's spread, we can identify deficiencies in decision-making, resource allocation, and firefighting methods during the rescue process, propose effective solutions, and minimize losses through scientific means. However, due to current limitations in simulation methods and computing capabilities, the combined simulation of forest fire spread and rescue efforts remains challenging. Summary of the Invention

[0004] The present invention provides a behavior tree-based rescue scene simulation method, device and equipment to solve the technical problems in related technologies of low simulation level of disaster rescue scenes and poor disaster prevention capabilities.

[0005] In a first aspect, an embodiment of the present invention provides a rescue scenario simulation method based on a behavior tree, the method comprising:

[0006] Analyze the rescue scenario that needs to be simulated to obtain the scenario information of the rescue scenario;

[0007] Determine a scenario computing model based on the scenario information, the computing model being used to simulate a state of a rescue scenario according to the scenario information;

[0008] Generate multiple disaster-causing agent agents and multiple rescue force agent agents based on scenario information and computational models;

[0009] Configure behavior trees for multiple disaster-causing agents and multiple rescue force agents, and simulate rescue scenarios.

[0010] In one possible implementation, in the method provided by an embodiment of the present invention, the calculation model includes a spreading model, a pathfinding model, a mobility model, and a rescue model, wherein:

[0011] The spread model is used to simulate the spread of disasters in rescue scenarios;

[0012] The pathfinding model is used to simulate the path planning of each rescue force agent in the rescue scenario;

[0013] The mobility model is used to simulate the position update of each rescue force agent in the rescue scenario;

[0014] The rescue model is used to simulate the spread of disasters in rescue scenarios under the action of multiple rescue force agents.

[0015] In one possible implementation, the method provided in an embodiment of the present invention generates multiple disaster-causing agent agents and multiple rescue force agent agents based on scenario information and a computational model, including:

[0016] Generate multiple disaster-causing agent agents based on the spreading model and rescue model;

[0017] Generate multiple rescue force agents based on the pathfinding model, mobility model and rescue model.

[0018] In one possible implementation, in the method provided in an embodiment of the present invention, after generating multiple disaster-causing factor agents and multiple rescue force agents based on scenario information and a computing model, the method further includes:

[0019] Configure corresponding attribute parameters for each disaster factor agent and each rescue force agent.

[0020] In one possible implementation, in the method provided by an embodiment of the present invention, configuring behavior trees for multiple disaster-causing agent agents and multiple rescue force agent agents includes:

[0021] According to the pre-set behavior tree decision process, a control node is created for each disaster-causing agent and each rescue force agent. The control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes.

[0022] Enter behavior and node parameters for each control node;

[0023] Set a corresponding action node for each control node so that each control node corresponds to a behavior.

[0024] In one possible implementation, the method provided in the embodiment of the present invention further includes:

[0025] Set the step size for the rescue scene simulation so that the step size increases by one each time the behavior tree is run.

[0026] In one possible implementation, in the method provided by an embodiment of the present invention, after configuring behavior trees for multiple hazard-causing agents and multiple rescue force agents and performing rescue scenario simulations, the method further includes:

[0027] Get the simulation results of the rescue scenario;

[0028] Based on the simulation results, the attribute parameters corresponding to each disaster factor agent and each rescue force agent are updated.

[0029] In a second aspect, an embodiment of the present invention provides a rescue scenario simulation device based on a behavior tree, characterized by comprising:

[0030] An analysis unit, configured to analyze a rescue scenario to be simulated and obtain scene information of the rescue scenario;

[0031] a determination unit, configured to determine a scenario calculation model based on the scenario information, the calculation model being configured to simulate a state of a rescue scenario according to the scenario information;

[0032] A generation unit, configured to generate multiple disaster-causing factor agents and multiple rescue force agents based on scenario information and a computational model;

[0033] The configuration unit is used to configure behavior trees for multiple disaster-causing agent agents and multiple rescue force agents, and simulate rescue scenarios.

[0034] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the computing model includes a spreading model, a pathfinding model, a mobility model, and a rescue model, wherein:

[0035] The spread model is used to simulate the spread of disasters in rescue scenarios;

[0036] The pathfinding model is used to simulate the path planning of each rescue force agent in the rescue scenario;

[0037] The mobility model is used to simulate the position update of each rescue force agent in the rescue scenario;

[0038] The rescue model is used to simulate the spread of disasters in rescue scenarios under the action of multiple rescue force agents.

[0039] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the generating unit is specifically configured to:

[0040] Generate multiple disaster-causing agent agents based on the spreading model and rescue model;

[0041] Generate multiple rescue force agents based on the pathfinding model, mobility model and rescue model.

[0042] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the generating unit is further configured to:

[0043] Configure corresponding attribute parameters for each disaster factor agent and each rescue force agent.

[0044] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the configuration unit is specifically configured to:

[0045] According to the pre-set behavior tree decision process, a control node is created for each disaster-causing agent and each rescue force agent. The control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes.

[0046] Enter behavior and node parameters for each control node;

[0047] Set a corresponding action node for each control node so that each control node corresponds to a behavior.

[0048] In one possible implementation, in the apparatus provided by the embodiment of the present invention, the configuration unit is further configured to:

[0049] Set the step size for the rescue scene simulation so that the step size increases by one each time the behavior tree is run.

[0050] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the configuration unit is specifically configured to:

[0051] Get the simulation results of the rescue scenario;

[0052] Based on the simulation results, the attribute parameters corresponding to each disaster factor agent and each rescue force agent are updated.

[0053] In a third aspect, an embodiment of the present invention provides an electronic device comprising: at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method provided in the first aspect of the embodiment of the present invention.

[0054] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having computer program instructions stored thereon, which implements the method provided in the first aspect of the embodiment of the present invention when the computer program instructions are executed by a processor.

[0055] In an embodiment of the present invention, the rescue scenario to be simulated is first analyzed to obtain the scenario information. Based on this scenario information, a computational model for the scenario is then determined. The computational model is used to simulate the state of the rescue scenario based on the scenario information. Multiple disaster-causing agent agents and multiple rescue force agents are then generated based on the scenario information and the computational model. Finally, behavior trees are configured for the multiple disaster-causing agent agents and multiple rescue force agents, and the rescue scenario is simulated. Compared to related technologies, this approach addresses the issues of low-level simulation of disaster rescue scenarios and poor disaster prevention capabilities. By using computational models and creating agents, and utilizing behavior trees to simulate disaster scenarios in real time, the agents are able to independently determine the appropriate actions to execute in their respective spatiotemporal environments based on their own behavior trees. This approach, while being scalable, achieves a highly realistic simulation of the entire rescue process. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flowchart of a rescue scenario simulation method based on a behavior tree provided by an embodiment of the present invention;

[0057] Figure 2 A schematic diagram of a specific process of a rescue scenario simulation method based on a behavior tree provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of a route planned by a pathfinding model according to an embodiment of the present invention;

[0059] Figure 4 A schematic diagram of the mobility model operation logic provided by an embodiment of the present invention;

[0060] Figure 5 A schematic diagram of the behavior tree structure of a fire brigade agent provided in an embodiment of the present invention;

[0061] Figure 6 A schematic diagram of the behavior tree structure of a forest fire scene agent provided by an embodiment of the present invention;

[0062] Figure 7 A schematic diagram of the agent logic call provided by an embodiment of the present invention;

[0063] Figure 8 A schematic diagram of a forest fire scene simulation provided by an embodiment of the present invention;

[0064] Figure 9 A schematic structural diagram of a rescue scenario simulation device based on a behavior tree provided by an embodiment of the present invention;

[0065] Figure 10 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] To make the objectives, technical solutions, and advantages of the present invention more apparent, the present invention will be further described in detail below with reference to the accompanying drawings. It should be understood that the embodiments described herein are merely some, rather than all, of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0067] The following are explanations of some of the words that appear in the text:

[0068] 1. In the embodiments of the present invention, the term "and / or" describes the association relationship between associated objects, indicating that three possible relationships exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0069] 2. Behavior tree is a graphical modeling language with a tree structure composed of behavior nodes. After using behavior tree, intelligent agents can increase their information perception of complex environments.

[0070] 3. Control nodes are internal nodes in a behavior tree. They define how to traverse their child nodes. As a parent node, a control node can have multiple child nodes, which can be either execution nodes or control nodes. Control nodes can be divided into three categories: sequence nodes, fallback nodes, and parallel nodes.

[0071] 4. Condition Node: One of the execution nodes at the end of the behavior tree. This node is the judgment made by the agent based on its own attributes or the attributes of other agents. Each condition node is a logical operation expression.

[0072] 5. The action node (Reaction) is another execution node at the end of the behavior tree. This node is the trigger for the agent to call the behavior model based on its own properties or the properties of other agents. Each action node is a set of input parameters for the behavior model.

[0073] Forest fire is the most serious type of forest disaster. It is sudden, destructive, wide-ranging, difficult to control, and difficult to extinguish and rescue. Once it occurs, it will cause huge losses to forest resources and seriously affect people's lives and property safety.

[0074] A large proportion of forest fires are caused by untimely or inappropriate firefighting plans, which miss the optimal time to extinguish a small fire, leading to its development into a large-scale blaze. During firefighting, humans, the environment, and the fire form a system. Therefore, by simulating human behavior during a forest fire's spread, we can identify deficiencies in decision-making, resource allocation, and firefighting methods during the rescue process, propose effective solutions, and minimize losses through scientific means. However, due to current limitations in simulation methods and computing capabilities, the combined simulation of forest fire spread and rescue efforts remains challenging.

[0075] The present technical solution provides a rescue scenario simulation method based on a behavior tree. The following describes the rescue scenario simulation method, device, and equipment based on a behavior tree in more detail with reference to the accompanying drawings and embodiments.

[0076] The embodiment of the present invention provides a rescue scenario simulation method based on behavior tree, such as Figure 1 Shown, including:

[0077] Step S101: Analyze the rescue scene to be simulated to obtain scene information of the rescue scene.

[0078] During specific implementation, the scene is analyzed to obtain scene information of the scene, which includes information in multiple dimensions such as geographical factors, environmental factors, weather factors, human factors, and traffic factors.

[0079] Step S102: determining a calculation model of the scene based on the scene information.

[0080] During specific implementation, a computing model of the scene is determined based on the scene information. The computing model is used to simulate the state of the rescue scene according to the scene information. The computing model includes a spread model, a pathfinding model, a mobility model and a rescue model. The spread model is used to simulate the spread of disasters in the rescue scene; the pathfinding model is used to simulate the path planning of each rescue force agent in the rescue scene; the mobility model is used to simulate the position update of each rescue force agent in the rescue scene; and the rescue model is used to simulate the spread of disasters in the rescue scene under the action of multiple rescue force agents.

[0081] Step S103: Generate multiple disaster-causing factor agents and multiple rescue force agents based on the scenario information and the calculation model.

[0082] During specific implementation, multiple disaster-causing factor agents are generated based on the spread model and the rescue model, and multiple rescue force agents are generated based on the path-finding model, the mobility model and the rescue model.

[0083] After generation, corresponding attribute parameters are configured for each disaster-causing factor agent and each rescue force agent.

[0084] Step S104: configure behavior trees for multiple disaster-causing agent agents and multiple rescue force agent agents, and simulate rescue scenarios.

[0085] In implementation, a control node is created for each hazard agent and each rescue force agent based on a pre-defined behavior tree decision process. These control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes. Behaviors and node parameters are then entered for each control node, and corresponding action nodes are assigned to each control node, ensuring that each control node corresponds to a behavior.

[0086] You can also set a step size for the rescue scenario simulation, increasing it by one each time the behavior tree is run. After each simulation, you can obtain the rescue scenario's simulation results and, based on these results, update the corresponding attribute parameters for each hazard agent and each rescue force agent to ensure the simulation's timeliness.

[0087] like Figure 2 As shown, taking the forest fire spreading and rescue scenario as an example, the rescue scenario simulation method based on behavior tree provided by the embodiment of the present invention is specifically described:

[0088] Step S201: Analyze the rescue scenario to be simulated to obtain scenario information of the rescue scenario.

[0089] During specific implementation, the scene is analyzed to obtain scene information of the scene, which includes information in multiple dimensions such as geographical factors, environmental factors, weather factors, human factors, and traffic factors.

[0090] Step S202: determining a calculation model of the scene based on the scene information.

[0091] During specific implementation, a computing model of the scene is determined based on the scene information. The computing model is used to simulate the state of the rescue scene according to the scene information. The computing model includes a spread model, a pathfinding model, a mobility model and a rescue model. The spread model is used to simulate the spread of disasters in the rescue scene; the pathfinding model is used to simulate the path planning of each rescue force agent in the rescue scene; the mobility model is used to simulate the position update of each rescue force agent in the rescue scene; and the rescue model is used to simulate the spread of disasters in the rescue scene under the action of multiple rescue force agents.

[0092] Specifically, the models used by the behavior tree-driven intelligent agent in the present invention include the spreading model, the pathfinding model, the movement model, and the rescue model. The following briefly introduces the model content.

[0093] (1) Spread model

[0094] In this paper, a forest fire spread model combining the Rothermel, elliptic, and Huygens models is used to analyze the forest fire spread process. The Rothermel model uses the principle of energy conservation and integrates slope wind speed and combustible material information to calculate the maximum forest fire spread speed. The elliptic model deduces the fire spread shape and estimates the speed of the fire spreading in different directions around it based on wind direction and maximum spread speed. The Huygens model regards the fire point as a wave source and the forest fire spread process as the propagation process of the wave surface. Combining these three models can calculate the changing process of the fire point spreading speed in different directions, thereby achieving the effect of forest fire spread.

[0095] (2) Pathfinding model

[0096] In forest fire search and rescue scenarios, both firefighters and fire trucks need to perform path planning. In the dangerous areas of the complex forest fire environment, they need to determine a better route to reach the designated fire extinguishing target location. However, in the actual fire fighting process, it is often difficult to quickly find a reasonable route due to the complexity of the environment and the uncertainty of the development of the disaster. However, in the simulation of forest fire rescue, the intelligent agent can obtain detailed information about the forest fire scene in a similar way to the "God's perspective" and quickly give the corresponding route to avoid the fire area through the designed path-finding algorithm. In the present invention, an improved particle swarm automatic risk avoidance algorithm is adopted to quickly search for a reasonable path that effectively avoids the dynamically changing forest fire environment area by generating a path population. The core formula of the algorithm is as follows:

[0097]

[0098]

[0099]

[0100] in, Update weight for the current moment speed, Update weight for the previous moment's velocity, is a random number from (0,1), is the optimal position in its own population, is the optimal position among all populations, is the current position of the particle, is the self-awareness factor at the current moment, is the social cognitive factor at the current moment, d is the current iteration number, D is the total iteration number, 、 、 are the minimum values ​​of weight, self-cognition factor, and social cognition factor, respectively.

[0101] Based on the above algorithm, the agent can find a reasonable route during the spread of forest fire. The planning effect is as follows: Figure 3 shown.

[0102] (3) Mobile model

[0103] The model updates the position of the agent in the forest fire rescue scenario based on the agent's own coordinates and the route calculated by the pathfinding model. The operation logic of this process is as follows: Figure 4 shown.

[0104] The behavior tree-driven intelligent agent calls the movement model in each loop to realize the movement process of the intelligent agent in the forest fire rescue scenario.

[0105] (4) Rescue model

[0106] The rescue model performs calculations based on the spread model, giving different agents different fire extinguishing calculation formulas to achieve the desired fire extinguishing effect. For example, the fire brigade agent uses its own fire extinguishing power as a benchmark, interacts with the fire scene environment agent, and calculates the fire extinguishing process based on the fire brigade agent's own fire extinguishing power attributes and fire scene intensity attributes; the fire truck agent calculates the fire extinguishing process based on its own water volume and water spraying radius and fire scene intensity attributes; the firefighting aircraft agent calculates the fire extinguishing process based on its own flight altitude and water volume carried and fire scene intensity attributes. The fire scene environment agent updates in real time during the fire extinguishing process. The core formulas for the fire extinguishing process calculations of the three types of agents are as follows:

[0107]

[0108] in, is the fire intensity of a cell. When it is 0, it means it is extinguished. For the fire fighting force of personnel, For the physical condition of personnel, Fire extinguishing capability for vehicles, The aircraft's flight altitude.

[0109] Step S203: Generate multiple disaster-causing factor agents and multiple rescue force agents based on the scenario information and the calculation model.

[0110] During specific implementation, multiple disaster-causing factor agents are generated based on the spread model and the rescue model, and multiple rescue force agents are generated based on the path-finding model, the mobility model and the rescue model.

[0111] In the present invention, two types of intelligent agents, disaster-causing factors and rescue forces, are established by taking the forest fire fighting scenario as an example. The disaster-causing factor is the forest fire scene, and the rescue forces include fire brigade, fire truck and fire aircraft intelligent agents. The process of constructing the rescue force intelligent agent is generally similar, and a detailed introduction is given by taking the fire brigade intelligent agent construction as an example.

[0112] Due to the complexity of forest fire rescue operations, the construction of behavior trees for various agents is complex. Therefore, the behavior tree data flow must adhere to a series of standard specifications to ensure the smooth operation of the simulation. In this paper, the standardization of the data flow of agent behavior trees mainly includes two aspects: agent attribute types and attribute value ranges.

[0113] Agent attribute type

[0114] Each type of agent has its own attributes. These attributes reflect the state of the agent at a specific time and can be affected by the interactions between agents. Therefore, before building an agent's behavior tree, it is necessary to define the attributes and types contained in each behavior and interaction process in the agent. In this invention, the following agent attribute types are specified:

[0115]

[0116] Attribute value range

[0117] After assigning agent attributes, their value ranges need to be constrained to prevent attribute overflows during the behavior tree's driving, which could lead to illogical behavior during the forest fire fighting simulation. For example, the firefighter agent's health must be non-negative, the aircraft's water load must be non-negative, and the fire truck agent's speed must be within a reasonable range. This ensures that the agent's attribute values ​​remain within a reasonable range when driven by the behavior tree, further ensuring that the behavior tree can operate without errors over a long period of time.

[0118] First, the agent is given the name "Fire Brigade." This serves as a unified template for this type of agent. All subsequently generated Fire Brigade entities are based on this template. Furthermore, corresponding attributes are added to the Fire Brigade agent, including location, travel speed, travel path, and firefighting tools. When adding attributes, the attribute name, attribute type, attribute unit, default value, and range are required to ensure smooth invocation of the behavior model. The process for creating attributes for the Forest Fire Scene agent is similar. The following table lists the attributes for the Fire Brigade agent and the Forest Fire Scene agent in this invention.

[0119]

[0120]

[0121] Step S204: configuring corresponding attribute parameters for each disaster-causing factor agent and each rescue force agent.

[0122] In the specific implementation, after setting the properties of the fire brigade and forest fire scene agents, the behavior models to be executed are configured in combination with the scenario requirements.

[0123] The input and output parameters of the fire brigade agent behavior are as follows:

[0124]

[0125] The five behaviors of the Forest Fire Agent, with the models and input and output parameters used for each behavior, are as follows:

[0126]

[0127] Based on the agent behavior triggers configured above, a series of behavior triggers can be further coupled to the agent behavior tree, ensuring that the attribute type and value range are reasonable.

[0128] Step S205 , configuring behavior trees for multiple disaster-causing agent agents and multiple rescue force agents, and performing rescue scenario simulations.

[0129] In implementation, a control node is created for each hazard agent and each rescue force agent based on a pre-defined behavior tree decision process. These control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes. Behaviors and node parameters are then entered for each control node, and corresponding action nodes are assigned to each control node, ensuring that each control node corresponds to a behavior.

[0130] You can also set a step size for the rescue scene simulation so that the step size increases by one each time the behavior tree is run.

[0131] The specific creation process is as follows:

[0132] 1. Create the overall control node structure based on the behavior tree decision process, including forward nodes, alternative nodes, parallel nodes, and conditional nodes;

[0133] 2. Configure the input behavior and parameters of each node, mainly for the triggering of conditional nodes. The required input includes the agent's own properties, the required properties of other agents, and trigger instructions.

[0134] 3. Configure the action nodes and select the corresponding behavior for each action node.

[0135] Specifically, for the behavior tree of the fire brigade agent, first, starting from the decision-making plan to be executed by the fire brigade agent, in the embodiment of the present invention, the fire brigade agent is explained as having two behavior modes of movement and fire extinguishing. At the bottom of the behavior tree structure, the alternative node is used as the root node, and two child nodes are branched out below the alternative node, which serve as subtrees for realizing the two behavior modes of movement and fire extinguishing respectively; secondly, for the two subtrees of movement and fire extinguishing, considering the continuity of their execution process, the sequential node is selected as the root node of the two subtrees. In addition, in order to ensure the independence and uniqueness of the subtree execution, the first leaf node executed below the sequential node is set as a conditional node. The conditional node is used to judge the instructions received by the agent. If its instructions do not conform to the original design function of the subtree, the action node configured under the subtree will not be executed; finally, according to the above-mentioned agent behavior list, action nodes and conditional nodes are configured for all subtrees. For the movement subtree, the configured action behaviors are: state modification behavior, path finding behavior, and travel behavior; for the fire extinguishing subtree, the configured actions are: state modification behavior and fire extinguishing behavior.

[0136] For the forest fire agent behavior tree, since its decision-making process is relatively simple, it only needs to calculate the spread of the fire and its related parameters. First, since the forest fire agent behavior tree does not require human-generated triggering, its root node is a sequence node, and a series of behavior nodes are located under the sequence node. Second, based on the overall functionality of the forest fire agent, its behavior is divided into two parts. The first part is the spread behavior, which is integrated into the spread model and can be used after configuring the aforementioned agent properties and interface input and output parameters. The second part is the behavior that calculates key indicators based on the spread process, including the current burned area, the current extinguished area, and the current burned area. Finally, these behaviors are sorted and added as leaf nodes of the forest fire agent behavior tree by adding control nodes. In this example, the second part of the behavior is controlled by the candidate node and the sequence node. Only after the spread behavior is successfully executed can the series of calculation behaviors in the second part be executed sequentially through the sequence node.

[0137] like Figure 5 and Figure 6 The following are examples of the behavior tree structure of the fire brigade agent and the forest fire scene agent, respectively.

[0138] After creating a behavior tree-based agent template, we need to further construct the entire forest fire spread and rescue scenario so that the various agents can interact and evolve within the specific scenario. Specifically, after the agent template is created, we need to generate agent instances with different attribute values ​​based on the template.

[0139] In this invention, while ensuring efficient model invocation, each agent in a scenario runs a behavior tree at a fixed frequency. Each time a behavior tree is executed, i.e., after one time step, the agent's attribute clock increments, and the behavior trigger at the root node, determined by the behavior tree's rules, is executed. The information in the behavior trigger is then passed to the model to be invoked, and the agent's attributes are updated after the model's return value is calculated. This repetitive cycle continuously updates the agent's attributes and behaviors, achieving a behavior-tree-driven agent evolution process.

[0140] Step S206: obtain the simulation results of the rescue scenario, and update the attribute parameters corresponding to each disaster factor agent and each rescue force agent based on the simulation results.

[0141] In practice, when using behavior trees to drive intelligent agents, to ensure effective behavior tree execution, model parameters must be checked before calling the model. Furthermore, the behavior model calling process is divided into two categories based on varying decision complexity: one is based on calling the local intelligent agent's built-in model, and the other is based on cross-domain model calling.

[0142] In the process of executing each action node of the agent, the internal attributes of the agent are called as the input of the behavior model. In this process, the calling attribute parameters need to be checked, including attribute null value check, attribute type check, and attribute value range check. If the called attribute does not meet the requirements during the check, the corresponding action model will not be called. The logical process is as follows Figure 7 shown.

[0143] In the embodiments of the present invention, the most critical aspect is the construction of behavior trees. As a tool for constructing complex behaviors of non-human characters, the rules for each node in a behavior tree are relatively mature. However, there is still room for improvement in its application to intelligent agents. Based on this, new rules are further specified for the construction of behavior trees for intelligent agents, including completeness rules, connectivity rules, coupling rules, and sequence rules.

[0144] (1) Completeness rules for agent behavior trees

[0145] During the Behavior Tree construction process, the integrity of the agent's functionality must be ensured. Early in the construction process, the agent's intended functions must be organized into a function-behavior list. Only after configuring the corresponding properties and behaviors can selections be made from the Behavior Tree's root node. Furthermore, each branch of the constructed Behavior Tree must have a corresponding root node, and this root node must be an action node, not a control node.

[0146] (2) Connectivity rules of agent behavior tree

[0147] When executing a behavior tree, first check the connectivity of each behavior tree node. In particular, ensure that each control node has a corresponding action node under it, and that the behavior of this action node is built into the agent, or that it can return a value when accessing remote services. Otherwise, if an exception occurs in a behavior node, the entire behavior tree will cease to run.

[0148] (3) Coupling rules of agent behavior trees

[0149] The input parameters of the models called in the behavior tree are not just the attributes of the agent itself. In certain situations, when interacting with other agents, the agent may need to obtain certain attributes of other agents and couple them to its own behavior tree to execute decisions. To this end, the present invention develops corresponding coupling rules during the behavior tree construction process to obtain certain attributes of other agents.

[0150] (4) Sequential rules of agent behavior trees

[0151] During the operation of multiple agents, if there are situations where the agents reference each other's attributes, a sequence problem will arise. Therefore, to solve this problem, the running order rules of the behavior trees between the agents are specified. During the agent initialization process, the agent is given a running order identification code. Based on this identification code, the recursive loop caused by mutual references between agents can be broken, ensuring the reasonable and orderly process of each agent calling the behavior tree.

[0152] Based on the above process, the process simulation under the conditions of mutual influence between the forest fire spread process and the fire extinguishing process of the rescue force in the forest fire scene can be realized, such as Figure 8 As shown. Figure 8 The forest fire scenario shown involves four types of agents: the forest fire scene agent, the fire rescue team agent, the fire rescue vehicle agent, and the firefighting aircraft agent. Each agent has its own behavior and behavior tree. Throughout the simulation, each agent instance invokes the required internal or external models based on the behavior triggering rules in its behavior tree, taking into account the changing conditions of its spatiotemporal environment and related attributes. This enables the on-demand invocation of multiple models within a single forest fire simulation scenario, ultimately resulting in a complete simulation of the entire forest fire scenario.

[0153] like Figure 9 As shown, based on the same inventive concept as the behavior tree-based rescue scenario simulation method, the present invention also provides a behavior tree-based rescue scenario simulation device, comprising:

[0154] An analysis unit 901 is configured to analyze a rescue scenario to be simulated and obtain scene information of the rescue scenario;

[0155] A determination unit 902 is configured to determine a scenario calculation model based on the scenario information, wherein the calculation model is configured to simulate a state of a rescue scenario according to the scenario information;

[0156] A generation unit 903 is configured to generate multiple disaster-causing factor agents and multiple rescue force agents based on scenario information and a calculation model;

[0157] The configuration unit 904 is used to configure behavior trees for multiple disaster-causing factor agents and multiple rescue force agents, and simulate rescue scenarios.

[0158] In one possible implementation, in the apparatus provided by an embodiment of the present invention, the computing model includes a spreading model, a pathfinding model, a mobility model, and a rescue model, wherein:

[0159] The spread model is used to simulate the spread of disasters in rescue scenarios;

[0160] The pathfinding model is used to simulate the path planning of each rescue force agent in the rescue scenario;

[0161] The mobility model is used to simulate the position update of each rescue force agent in the rescue scenario;

[0162] The rescue model is used to simulate the spread of disasters in rescue scenarios under the action of multiple rescue force agents.

[0163] In one possible implementation, in the apparatus provided by the embodiment of the present invention, the generating unit 903 is specifically configured to:

[0164] Generate multiple disaster-causing agent agents based on the spreading model and rescue model;

[0165] Generate multiple rescue force agents based on the pathfinding model, mobility model and rescue model.

[0166] In a possible implementation, in the apparatus provided by the embodiment of the present invention, the generating unit 903 is further configured to:

[0167] Configure corresponding attribute parameters for each disaster factor agent and each rescue force agent.

[0168] In one possible implementation, in the apparatus provided by the embodiment of the present invention, the configuration unit 904 is specifically configured to:

[0169] According to the pre-set behavior tree decision process, a control node is created for each disaster-causing agent and each rescue force agent. The control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes.

[0170] Enter behavior and node parameters for each control node;

[0171] Set a corresponding action node for each control node so that each control node corresponds to a behavior.

[0172] In a possible implementation, in the apparatus provided by the embodiment of the present invention, the configuration unit 904 is further configured to:

[0173] Set the step size for the rescue scene simulation so that the step size increases by one each time the behavior tree is run.

[0174] In one possible implementation, in the apparatus provided by the embodiment of the present invention, the configuration unit 904 is specifically configured to:

[0175] Get the simulation results of the rescue scenario;

[0176] Based on the simulation results, the attribute parameters corresponding to each disaster factor agent and each rescue force agent are updated.

[0177] like Figure 10 As shown, electronic device 1000 may include a processing device (e.g., a central processing unit, graphics processing unit, etc.) 1001, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 1002 or programs loaded from storage device 1008 into random access memory (RAM) 1003 to implement the behavior tree-based rescue scenario simulation method according to the embodiments of the present disclosure. RAM 1003 also stores various programs and data required for the operation of electronic device 1000. Processing device 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. An input / output (I / O) interface 1005 is also connected to bus 1004.

[0178] Typically, the following devices may be connected to the I / O interface 1005: an input device 1006 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1007 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1008 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the electronic device 1000 to communicate with other devices wirelessly or by wire to exchange data. Figure 10 The electronic device 1000 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0179] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart, thereby implementing the voice control method as described above. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 1009, or installed from the storage device 1008, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0180] It should be noted that the computer-readable medium described above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wire, optical cable, RF (radio frequency), or any suitable combination thereof.

[0181] In some embodiments, the client and server can communicate using any currently known or later developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or later developed network.

[0182] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0183] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device:

[0184] Analyze the rescue scenario that needs to be simulated to obtain the scenario information of the rescue scenario;

[0185] Determine a scenario computing model based on the scenario information, the computing model being used to simulate a state of a rescue scenario according to the scenario information;

[0186] Generate multiple disaster-causing agent agents and multiple rescue force agent agents based on scenario information and computational models;

[0187] Configure behavior trees for multiple disaster-causing agents and multiple rescue force agents, and simulate rescue scenarios.

[0188] Optionally, when the above one or more programs are executed by the electronic device, the electronic device may also execute other steps described in the above embodiments.

[0189] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0190] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0191] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit does not necessarily limit the unit itself.

[0192] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0193] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0194] In an embodiment of the present invention, the rescue scenario to be simulated is first analyzed to obtain the scenario information. Based on this scenario information, a computational model for the scenario is then determined. The computational model is used to simulate the state of the rescue scenario based on the scenario information. Multiple disaster-causing agent agents and multiple rescue force agents are then generated based on the scenario information and the computational model. Finally, behavior trees are configured for the multiple disaster-causing agent agents and multiple rescue force agents, and the rescue scenario is simulated. Compared to related technologies, this approach addresses the issues of low-level simulation of disaster rescue scenarios and poor disaster prevention capabilities. By using computational models and creating agents, and utilizing behavior trees to simulate disaster scenarios in real time, the agents are able to independently determine the appropriate actions to execute in their respective spatiotemporal environments based on their own behavior trees. This approach, while being scalable, achieves a highly realistic simulation of the entire rescue process.

[0195] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0196] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0197] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0198] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0199] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0200] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A rescue scenario simulation method based on behavior tree, characterized in that: include: Analyze the rescue scenario to be simulated to obtain scenario information of the rescue scenario; Determining a computing model of the scenario based on the scenario information, wherein the computing model is used to simulate a state of the rescue scenario according to the scenario information; The computational model includes a spreading model, a pathfinding model, a mobility model, and a rescue model, wherein: The spreading model is used to simulate the spreading of the disaster in the rescue scenario; The path-finding model is used to simulate the path planning of each rescue force agent in the rescue scenario; The mobility model is used to simulate the position update of each rescue force agent in the rescue scenario; The rescue model is used to simulate the spread of the disaster under the action of multiple rescue force agents in the rescue scenario; generating a plurality of disaster-causing factor agents based on the spreading model and the rescue model; generating the plurality of rescue force agents based on the pathfinding model, the mobility model, and the rescue model; Behavior trees are configured for the multiple disaster-causing factor agents and the multiple rescue force agents, and a simulation of the rescue scenario is performed.

2. The rescue scenario simulation method based on behavior tree according to claim 1 is characterized in that: After generating a plurality of disaster-causing factor agents and a plurality of rescue force agents based on the scenario information and the calculation model, the method further includes: Configure corresponding attribute parameters for each disaster factor agent and each rescue force agent.

3. The rescue scenario simulation method based on behavior tree according to claim 2 is characterized in that: The configuring behavior trees for the multiple disaster-causing agent agents and the multiple rescue force agent agents includes: According to the pre-set behavior tree decision process, a control node is created for each disaster-causing agent and each rescue force agent. The control nodes include forward nodes, alternative nodes, parallel nodes, and conditional nodes. inputting behavior and node parameters for each of the control nodes; A corresponding action node is set for each of the control nodes, so that each control node corresponds to the behavior.

4. The rescue scenario simulation method based on behavior tree according to claim 3 is characterized in that: The method further comprises: Set the step size for the rescue scenario simulation so that the step size increases by one each time the behavior tree is run.

5. The rescue scenario simulation method based on behavior tree according to claim 4 is characterized in that: After configuring behavior trees for the multiple disaster-causing agent agents and the multiple rescue force agent agents and performing the rescue scenario simulation, the method further includes: Obtaining a simulation result of the rescue scenario; Based on the simulation results, the attribute parameters corresponding to each disaster-causing factor agent and each rescue force agent are updated.

6. A rescue scene simulation device based on behavior tree, characterized in that: include: An analysis unit, configured to analyze a rescue scenario to be simulated and obtain scenario information of the rescue scenario; a determining unit, configured to determine a calculation model of the scenario based on the scenario information, wherein the calculation model is configured to simulate a state of the rescue scenario according to the scenario information; The computational model includes a spreading model, a pathfinding model, a mobility model, and a rescue model, wherein: The spreading model is used to simulate the spreading of the disaster in the rescue scenario; The path-finding model is used to simulate the path planning of each rescue force agent in the rescue scenario; The mobility model is used to simulate the position update of each rescue force agent in the rescue scenario; The rescue model is used to simulate the spread of the disaster under the action of multiple rescue force agents in the rescue scenario; a generating unit, configured to generate a plurality of hazard factor agents based on the spreading model and the rescue model; and generate a plurality of rescue force agents based on the pathfinding model, the mobility model, and the rescue model; A configuration unit is used to configure behavior trees for the multiple disaster-causing factor agents and the multiple rescue force agents, and to simulate the rescue scenario.

7. An electronic device, characterized in that: include: At least one processor, at least one memory, and computer program instructions stored in the memory, which implement the method according to any one of claims 1 to 5 when the computer program instructions are executed by the processor.

8. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images