A method and device for constructing a simulation entity task model

Abstract

The application discloses a kind of construction method and device of simulation entity task model, the method includes obtaining the n entity objects needing to execute task, n is positive integer;The n entity objects are processed, and simulation entity task model is obtained;The task action plan of entity is obtained;The task action plan of the entity is processed using the simulation entity task model, and task action plan execution result is obtained.The simulation entity task model of the application is independently constructed outside entity's task model, and corresponding task ability operation tool is provided, can set the behavior sequence of entity according to different simulation needs, reach the purpose of flexible setting task ability model.

Description

Technical Field

[0001] This invention relates to the field of physical task simulation technology, and in particular to a method and apparatus for constructing a simulated physical task model. Background Technology

[0002] In simulation application systems, entity models are a crucial component. By submitting event execution requests, these models, under the unified scheduling of the simulation engine, complete behavioral actions and information interactions. Corresponding to the physical world, the final behavioral performance of a simulated entity model is primarily determined by its basic capabilities and task capabilities. Basic capabilities are determined by the entity's attribute parameters and fundamental functions, while task capabilities mainly manifest as the planning and adjustment of the sequence of actions the entity takes to complete a target task.

[0003] Traditionally, simulation entity models couple the entity's basic capability model with the task capability model. In the simulation, the entity responds to external stimuli, and the task model mobilizes the entity's capability model to complete the corresponding task. This approach has the following problems: First, the complexity of model design and development is high. Model designers need a comprehensive understanding of the mechanism of the entity's physical action capabilities, as well as the state transition conditions and action decision-making rules under various external environments. Model developers need to implement a large number of conditional judgments and state transitions in the code. As the number of task types and factors increases, the difficulty of program implementation increases, leading to higher defects and failure rates, and even making implementation impossible. Second, in human-in-the-loop application modes, the simulated entity needs to respond to human decisions and adjustments. This requires prioritizing the simulated entity's response to human decisions and commands, further increasing the difficulty of model design and development. Third, with changes in weapon and equipment usage modes and tactical innovations, the task types and state transition conditions of weapon and equipment also change. The original designed and developed models did not consider new application modes, and existing model behaviors may no longer meet the new usage requirements, necessitating continuous upgrades or adjustments to the models based on changes in tactical applications. Modifying and upgrading based on an existing model leads to issues such as code refactoring and retesting, while completely redesigning and developing the model results in the form of duplicate work and the complete loss of previous work results. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and apparatus for constructing a simulation entity task model, which decouples the entity's basic capabilities and task capabilities, externally constructs the entity's task model, and provides corresponding task capability operation tools, which can set the entity's behavior sequence according to different simulation requirements, thereby achieving the purpose of flexibly setting the task capability model.

[0005] To address the aforementioned technical problems, a first aspect of this invention discloses a method for constructing a simulation entity task model, the method comprising:

[0006] S1, obtain n entity objects for which the task needs to be performed, where n is a positive integer;

[0007] S2, process the n entity objects to obtain the simulation entity task model;

[0008] S3, Obtain the entity's task action plan;

[0009] S4. Using the simulated entity task model, process the entity's task action plan to obtain the task action plan execution result;

[0010] S5, evaluate the execution results of the task action plan to obtain the task action plan evaluation results.

[0011] As an optional implementation, in the first aspect of the present invention, processing the n entity objects to obtain the simulation entity task model includes:

[0012] S21, Process the n entity objects to obtain n single entity object task models;

[0013] S22, process the n single entity object task models to obtain the simulation entity task model.

[0014] As an optional implementation, in the first aspect of the present invention, the step of processing the n entity objects to obtain n single entity object task models includes:

[0015] S211, Process the n entity objects to obtain the task types of the n entity objects;

[0016] S212, using a preset task manager, process the task types of the n entity objects to obtain a task queue of n entity objects;

[0017] S213, process the task queue of the n entity objects to obtain n single entity object task models.

[0018] As an optional implementation, in the first aspect of the present invention, the task queue expression for the n entity objects is:

[0019]

[0020] Where QNum is the total number of entity task queues in a single simulation scenario, n is the total number of entities in the scenario, m is the number of task types for the i-th entity, and Tij Let j be the task type of the i-th entity.

[0021] As an optional implementation, in the first aspect of the present invention, processing the task queues of the n entity objects to obtain n single entity object task models includes:

[0022] S2131, Using the task manager, the task types of n entity objects are arranged to obtain task planning information;

[0023] S2132, Monitor the task execution of the task planning information to obtain execution status information;

[0024] S2133, Adjust the execution status information to obtain n single entity object task models.

[0025] As an optional implementation, in the first aspect of the present invention, the simulated entity task model includes an entity task configuration subsystem, a task list maintenance subsystem, a task execution scheduling subsystem, a task status monitoring subsystem, and a task dynamic adjustment subsystem;

[0026] The entity task configuration subsystem is connected to the task list maintenance subsystem for setting single conditions or condition groups for individual task items.

[0027] The task list maintenance subsystem is data-connected to the task operation scheduling subsystem, and is used to provide a human-computer interaction operation entry point for combat task planning.

[0028] The task execution scheduling subsystem is connected to the task status monitoring subsystem and the task dynamic adjustment subsystem via data connection, and is used to provide secondary development interface specifications for the task model;

[0029] The task status monitoring subsystem is used to provide task execution status monitoring functions based on the task list;

[0030] The task dynamic adjustment subsystem is used to provide human-computer interaction functions for real-time adjustment of task execution strategies.

[0031] As an optional implementation, in the first aspect of the present invention, the step of processing the task action plan of the entity using the simulated entity task model to obtain the task action plan execution result includes:

[0032] S41, Obtain the entity's task action plan;

[0033] S42, using the task configuration subsystem, process the task action plan of the entity to obtain the task type and task sequence of each entity in the scenario;

[0034] S43, using the task list maintenance subsystem, start simulation for the task type and task sequence of each entity in the scenario to obtain task action plan information;

[0035] S44, Using the task status monitoring subsystem, monitor the task action plan information, and update the execution status of each entity based on the monitoring results to obtain entity update information;

[0036] S45, using the task dynamic adjustment subsystem, update the task action plan information according to the entity update information to obtain the task action plan execution result.

[0037] A second aspect of this invention discloses an apparatus for constructing a simulation entity task model, the apparatus comprising:

[0038] The entity object acquisition module is used to acquire n entity objects for which the task needs to be performed, where n is a positive integer;

[0039] The simulation entity task model construction module is used to process the n entity objects to obtain the simulation entity task model;

[0040] The task action plan acquisition module is used to acquire the task action plans of entities;

[0041] The task action plan execution module is used to process the task action plan of the simulated entity using the simulated entity task model, and obtain the task action plan execution result;

[0042] The evaluation module is used to evaluate the execution results of the task action plan and obtain the task action plan evaluation results.

[0043] As an optional implementation, in the second aspect of the present invention, the processing of the n entity objects to obtain the simulation entity task model includes:

[0044] S21, Process the n entity objects to obtain n single entity object task models;

[0045] S22, process the n single entity object task models to obtain the simulation entity task model.

[0046] As an optional implementation, in the second aspect of the present invention, the step of processing the n entity objects to obtain n single entity object task models includes:

[0047] S211, Process the n entity objects to obtain the task types of the n entity objects;

[0048] S212, using a preset task manager, process the task types of the n entity objects to obtain a task queue of n entity objects;

[0049] S213, process the task queue of the n entity objects to obtain n single entity object task models.

[0050] As an optional implementation, in the second aspect of the present invention, the task queue expression for the n entity objects is:

[0051]

[0052] Where QNum is the total number of entity task queues in a single simulation scenario, n is the total number of entities in the scenario, m is the number of task types for the i-th entity, and T ij Let j be the task type of the i-th entity.

[0053] As an optional implementation, in the second aspect of the present invention, processing the task queues of the n entity objects to obtain n single entity object task models includes:

[0054] S2131, Using the task manager, the task types of n entity objects are arranged to obtain task planning information;

[0055] S2132, Monitor the task execution of the task planning information to obtain execution status information;

[0056] S2133, Adjust the execution status information to obtain n single entity object task models.

[0057] As an optional implementation, in the second aspect of the present invention, the simulated entity task model includes an entity task configuration subsystem, a task list maintenance subsystem, a task execution scheduling subsystem, a task status monitoring subsystem, and a task dynamic adjustment subsystem;

[0058] The entity task configuration subsystem is connected to the task list maintenance subsystem for setting single conditions or condition groups for individual task items.

[0059] The task list maintenance subsystem is data-connected to the task operation scheduling subsystem, and is used to provide a human-computer interaction operation entry point for combat task planning.

[0060] The task execution scheduling subsystem is connected to the task status monitoring subsystem and the task dynamic adjustment subsystem via data connection, and is used to provide secondary development interface specifications for the task model;

[0061] The task status monitoring subsystem is used to provide task execution status monitoring functions based on the task list;

[0062] The task dynamic adjustment subsystem is used to provide human-computer interaction functions for real-time adjustment of task execution strategies.

[0063] As an optional implementation, in the second aspect of the present invention, the step of processing the task action plan of the entity using the simulated entity task model to obtain the task action plan execution result includes:

[0064] S41, Obtain the entity's task action plan;

[0065] S42, using the task configuration subsystem, process the task action plan of the entity to obtain the task type and task sequence of each entity in the scenario;

[0066] S43, using the task list maintenance subsystem, start simulation for the task type and task sequence of each entity in the scenario to obtain task action plan information;

[0067] S44, Using the task status monitoring subsystem, monitor the task action plan information, and update the execution status of each entity based on the monitoring results to obtain entity update information;

[0068] S45, using the task dynamic adjustment subsystem, update the task action plan information according to the entity update information to obtain the task action plan execution result.

[0069] A third aspect of the present invention discloses another apparatus for constructing a simulation entity task model, the apparatus comprising:

[0070] Memory containing executable program code;

[0071] A processor coupled to the memory;

[0072] The processor calls the executable program code stored in the memory to execute some or all of the steps in the method for constructing a simulation entity task model disclosed in the first aspect of the present invention.

[0073] The fourth aspect of the present invention discloses a computer-storable medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the method for constructing a simulation entity task model disclosed in the first aspect of the present invention.

[0074] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0075] This invention proposes a method for constructing a simulation entity task model. This method can construct the entity's task model and the sequence of tasks between entities on demand, flexibly and independently. After adopting this method, the entity's task model is decoupled from the basic capability model, which reduces the difficulty of entity model design and development, and can adapt to the development and changes of weapon and equipment application modes.

[0076] This invention proposes an implementation framework for simulated entity task models. Using this framework, task construction, task scheduling, task monitoring, and task adjustment of entity task models can be realized, thus meeting the requirements for the construction and use of simulated entity task models.

[0077] This invention clarifies the operational mechanism of decoupling the simulation entity model from the basic capability model. By adopting the methods of task scheduling queue and entity task type queue, it realizes the parallel scheduling and operation of multiple tasks of the simulation entity, and conforms to the spatiotemporal consistency of the real world.

[0078] This invention proposes a process for constructing and using a task model, implements related functions based on a simulation entity task model framework, and clarifies the process for using the task model in simulation. Attached Figure Description

[0079] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0080] Figure 1 This is a flowchart illustrating a method for constructing a simulation entity task model according to an embodiment of the present invention;

[0081] Figure 2 This is the principle of single entity object task model construction and execution disclosed in the embodiments of the present invention;

[0082] Figure 3 This is the principle of entity object task model construction and execution disclosed in the embodiments of the present invention;

[0083] Figure 4 It is composed of the entity task model framework disclosed in the embodiments of the present invention;

[0084] Figure 5 This is the entity task model construction and usage process disclosed in the embodiments of the present invention;

[0085] Figure 6 This is a schematic diagram of the structure of a device for constructing a simulation entity task model disclosed in an embodiment of the present invention;

[0086] Figure 7 This is a schematic diagram of another device for constructing a simulation entity task model disclosed in an embodiment of the present invention. Detailed Implementation

[0087] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0088] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0089] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0090] The method and apparatus provided by this invention create conditions for decoupling the task model and basic capability model of an entity, enabling the model to meet the needs of different scenarios and constantly evolving equipment usage methods. It allows for setting the model's action sequence according to anticipated tactics and strategies during the scenario editing phase, and also enables real-time adjustment of the model's actions during simulation operation, achieving the construction and adjustment of entity task capabilities in both human-absent and human-in-the-loop simulation application modes. Simultaneously, basic capability model designers and developers only need to abstract the physical capabilities inherent in the equipment entity based on its existing capabilities, without needing to understand the complex and diverse equipment usage rules, significantly reducing the complexity and difficulty of model development.

[0091] This invention discloses a method and apparatus for constructing a simulated entity task model. The method includes obtaining n entity objects (n being a positive integer) that need to perform tasks; processing the n entity objects to obtain a simulated entity task model; obtaining the entity's task action plan; and processing the entity's task action plan using the simulated entity task model to obtain the task action plan execution result. This invention externally constructs the entity's task model independently and provides corresponding task capability operation tools, enabling the setting of entity behavior sequences according to different simulation requirements, thus achieving the goal of flexibly setting the task capability model. Detailed descriptions follow.

[0092] Example 1

[0093] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for constructing a simulation entity task model according to an embodiment of the present invention. Figure 1 The method for constructing the simulated entity task model described herein is applied in the field of entity task simulation technology, and the embodiments of this invention are not limited thereto. Figure 1 As shown, the method for constructing this simulation entity task model may include the following operations:

[0094] S1, obtain n entity objects for which the task needs to be performed, where n is a positive integer;

[0095] S2, process the n entity objects to obtain the simulation entity task model;

[0096] S3, Obtain the entity's task action plan;

[0097] S4. Using the simulated entity task model, process the entity's task action plan to obtain the task action plan execution result;

[0098] S5, evaluate the execution results of the task action plan to obtain the task action plan evaluation results.

[0099] Optionally, processing the n entity objects to obtain the simulation entity task model includes:

[0100] S21, Process the n entity objects to obtain n single entity object task models;

[0101] S22, process the n single entity object task models to obtain the simulation entity task model.

[0102] Optionally, the step of processing the n entity objects to obtain n single entity object task models includes:

[0103] S211, Process the n entity objects to obtain the task types of the n entity objects;

[0104] S212, using a preset task manager, process the task types of the n entity objects to obtain a task queue of n entity objects;

[0105] S213, process the task queue of the n entity objects to obtain n single entity object task models.

[0106] Optionally, the task queue expression for the n entity objects is:

[0107]

[0108] Where QNum is the total number of entity task queues in a single simulation scenario, n is the total number of entities in the scenario, m is the number of task types for the i-th entity, and T ij Let j be the task type of the i-th entity.

[0109] Optionally, processing the task queues of the n entity objects to obtain n single entity object task models includes:

[0110] S2131, Using the task manager, the task types of n entity objects are arranged to obtain task planning information;

[0111] S2132, Monitor the task execution of the task planning information to obtain execution status information;

[0112] S2133, Adjust the execution status information to obtain n single entity object task models.

[0113] Optionally, the simulated entity task model includes an entity task configuration subsystem, a task list maintenance subsystem, a task execution scheduling subsystem, a task status monitoring subsystem, and a task dynamic adjustment subsystem;

[0114] The entity task configuration subsystem is connected to the task list maintenance subsystem for setting single conditions or condition groups for individual task items.

[0115] The task list maintenance subsystem is data-connected to the task operation scheduling subsystem, and is used to provide a human-computer interaction operation entry point for combat task planning.

[0116] The task execution scheduling subsystem is connected to the task status monitoring subsystem and the task dynamic adjustment subsystem via data connection, and is used to provide secondary development interface specifications for the task model;

[0117] The task status monitoring subsystem is used to provide task execution status monitoring functions based on the task list;

[0118] The task dynamic adjustment subsystem is used to provide human-computer interaction functions for real-time adjustment of task execution strategies.

[0119] Optionally, the step of processing the task action plan of the entity using the simulated entity task model to obtain the task action plan execution result includes:

[0120] S41, Obtain the entity's task action plan;

[0121] S42, using the task configuration subsystem, process the task action plan of the entity to obtain the task type and task sequence of each entity in the scenario;

[0122] S43, using the task list maintenance subsystem, start simulation for the task type and task sequence of each entity in the scenario to obtain task action plan information;

[0123] S44, Using the task status monitoring subsystem, monitor the task action plan information, and update the execution status of each entity based on the monitoring results to obtain entity update information;

[0124] S45, using the task dynamic adjustment subsystem, update the task action plan information according to the entity update information to obtain the task action plan execution result.

[0125] Optionally, the method for evaluating the execution results of the task action plan and obtaining the task action plan evaluation results is as follows:

[0126] Analyzing the execution results of the task action plan yields a set of task action plan objects A = {A1, A2, ..., A...} n} and B = {B1, B2, ..., B m The two sets of objects represent the friendly side and the opposing side, respectively. n is the number of friendly objects, and m is the number of opposing objects. For each object, an index set X = {X1, X2, ..., X} is set. d}, where d is the number of indicators, establish the original data matrix R = (r ij ) l·d , where r ij Let l represent the data size of the j-th indicator of the i-th object, where l = n + m.

[0127] For each indicator X i ={r′ ij}, i=1,2,…,d,j=1,2,…,l all have:

[0128]

[0129] E = XW e +b e

[0130] Among them, W e ,b e These are the weight parameters and bias parameters of the linear embedding layer in the Transformer's self-attention mechanism, respectively, both of which are learned through training.

[0131] Incorporate learnable positional encoding to distinguish the position Z of each object:

[0132] Z = E + P

[0133] Where P = {p i} represents the positional encoding of the object, p i =W e x i +b e ,x i The actual position of object i;

[0134] Calculate the first matrix Q:

[0135] Q = ZW Q

[0136] Second matrix K:

[0137] K = ZW K

[0138] Third matrix V:

[0139] V = ZW V

[0140] In this system, Q is used to determine the information needs of the current object, K is used to judge the key descriptions of all objects, and V is used to pass its own information to the previous object to judge possible changes in the situation. W... Q W K W V They are all training matrices obtained through learning.

[0141] Enter the attention mechanism and calculate the attention score A:

[0142]

[0143] In the formula, softmax is the softmax function, T is the transpose, and d k is a constant, which is a scaling factor used to avoid gradient vanishing and prevent the dot product value from becoming too large.

[0144] Calculate the attention-weighted result O:

[0145] O = Attention(Q,K,V)

[0146] Attention refers to the attention mechanism;

[0147] Input Transformer encoder TramsformerEncoder:

[0148] H = TramsformerEncoder(O)

[0149] Get H, and each row of H is h i All of them incorporate the object i's sensitive information about its surrounding environment.

[0150] Scoring of the execution of the task action plan:

[0151]

[0152] Among them, W o ,b o These are the weight and bias parameters of the fully connected layer of the Transformer, both learned through training, which yields the task action plan execution score s for object i. i This refers to the result of the execution of the task action plan.

[0153] As can be seen, this invention proposes a method for constructing a simulated entity task model. This method can flexibly and independently construct the task model of an entity and the sequence of tasks between entities on demand. Using this method, the entity's task model is decoupled from the basic capability model, reducing the difficulty of entity model design and development, and adapting to the evolving operational modes of weaponry. This invention proposes an implementation framework for simulated entity task models. Using this framework, task construction, task scheduling, task monitoring, and task adjustment of the entity task model can be realized, satisfying the construction and use of simulated entity task models. This invention elucidates the operational mechanism of the decoupling between the simulated entity model and the basic capability model, and uses task scheduling queues and entity task type queues to achieve parallel scheduling and operation of multiple simulated entity tasks, while conforming to the spatiotemporal consistency of the real world.

[0154] This invention proposes a process for constructing and using a task model, implements related functions based on a simulation entity task model framework, and clarifies the process for using the task model in simulation.

[0155] Example 2

[0156] The method and technology used in this embodiment are as follows: design an implementation framework for building entity task models, and implement services such as task building, task orchestration, task execution, task scheduling, task control, task adjustment, status query, and task termination of entity task models; distinguish between two dimensions, object entities and task capabilities, and build a task scheduling queue for simulation entities; during simulation operation, the task scheduler schedules entity tasks sequentially according to task triggering conditions; the entity calls the corresponding basic capability model to execute actions according to the action sequence set in the task; and completes all action sequences according to the requirements of the task model to achieve the overall task objective.

[0157] 1. Principles of Task Model Implementation

[0158] The main purpose of this invention is to decouple the basic capabilities and task capabilities of an entity model. By flexibly constructing and setting the task capability model of an entity, the basic capability model of the entity is driven to complete the predetermined entity task.

[0159] During the simulation preparation phase, task sequences are uniformly set for all entity objects that need to perform tasks. Different entity objects are distinguished, and their own task sequences are set. The same entity object may have multiple task types. Different entity objects are assigned their own tasks, and different entity objects possess different task types based on their capabilities. The task manager manages the tasks of all entities uniformly, distinguishing entity objects and constructing queues for the tasks to be executed by each entity object according to task type. In a single simulation scenario, the total number of queues constructed is:

[0160]

[0161] Where QNum is the total number of entity task queues in a single simulation scenario, n is the total number of entities in the scenario, m is the number of task types for the i-th entity, and T ij Let j be the task type of the i-th entity.

[0162] After the simulation begins, a task queue of different task types is constructed for each entity object. The number of task queues for a single entity object is equal to the number of task types for that entity object. Each task of an entity enters the entity task type queue sequentially according to a predefined execution sequence. Tasks of the same type are grouped into one queue, and tasks of different types are placed in different task queues. The basis for constructing entity task queues according to the task types of different entities is that an entity object cannot perform the same task action at the same time; it can only perform the same action at different times. However, an entity object can perform different actions at the same time.

[0163] During simulation, tasks are dequeued sequentially according to their execution conditions, driving the entity's basic capability model to perform relevant actions. Different entities can execute their respective task actions simultaneously, and the same entity can execute different types of task actions concurrently, achieving parallel execution of multiple tasks, consistent with the spatiotemporal logic of events in the real world. The principle of task model construction and execution for a single entity object is as follows: Figure 2 As shown, the principle of entity object task model construction and execution is as follows: Figure 3 As shown.

[0164] 2. Functional Components of the Task Model Framework

[0165] The entity task model framework includes modules for entity task configuration, task list maintenance, task execution scheduling, task status monitoring, and dynamic task adjustment. The entity task model framework is composed of the following: Figure 4 As shown.

[0166] (1) Entity task configuration

[0167] The entity task configuration module provides a secondary development extension interface specification for task execution conditions, featuring a visual task execution condition configuration function. It allows for setting single conditions or groups of conditions for individual task items. The task execution conditions include built-in basic spatiotemporal conditions, such as relative time, absolute time, and sequential sequences. It supports logical combination settings and commonly used logical operators, including AND, OR, NOT, greater than, and less than. It also provides event acquisition and publishing interfaces, enabling the acquisition of event information such as combat, damage, detection, communication, and command, enriching the task condition settings.

[0168] (2) Task list maintenance

[0169] The task list maintenance module provides a human-computer interaction interface for operational mission planning, supporting the editing of entity tasks based on Gantt charts. It allows for the setting of task sequences for each entity according to plan requirements, including selecting entity objects, creating new tasks, selecting task types, configuring task execution conditions, adjusting task sequences, and setting task branch relationships.

[0170] (3) Task execution scheduling

[0171] The task execution and scheduling module provides a secondary development interface specification for the task model framework, enabling it to accept simulation engine execution control processing, including data loading, execution preparation, event submission and execution, and resource cleanup. Through the interfaces provided by the task execution and scheduling module, logical isolation is achieved between task sequences and capability-based task execution. The scheduler checks the execution conditions of each task to be executed and distributes the parameters of tasks that meet the conditions to the entity's basic capability model to execute the corresponding actions. It features dynamic task condition checking capabilities, enabling control of the task execution sequence according to task conditions or condition groups, and supports parallel scheduling of multiple tasks.

[0172] (4) Task status monitoring

[0173] The task status monitoring module provides task execution status monitoring functionality based on a task list. It can build a full sequence of task lists according to task type and monitor the status of each task in the list, including scheduling, execution, cancellation, termination, failure, and success. It provides task execution status data collection and statistics, facilitating real-time monitoring of the execution of task planning results.

[0174] (5) Dynamic adjustment of tasks

[0175] The mission dynamic adjustment module provides situational "human-in-the-loop" combat mission dynamic intervention function, and provides human-computer interaction for real-time adjustment of mission execution strategies, including adding missions, canceling missions, modifying mission parameters and other operations. It also records mission adjustment operation records in real time, which facilitates review and research on the adjustment and optimization of mission actions.

[0176] 3. Implementation of Task Model Framework Technology

[0177] The task model framework proposed in this invention provides a systematic set of interface functions for task model construction, management, and operation, which can meet the functions of entity task construction, entity task model scheduling, entity task execution status reporting, entity task execution status query, and entity task adjustment, as shown in Table 1.

[0178] Table 1. Implementation functions of the simulation entity task model framework

[0179]

[0180]

[0181]

[0182] 4. Entity Task Model Construction and Use

[0183] Based on the entity task model framework and function proposed in this invention, it is possible to realize task creation, entity task scheduling and execution, entity task execution status monitoring, and real-time adjustment of entity tasks for all entities in the simulation scenario. The main usage flow and methods are as follows: Figure 5 As shown:

[0184] (1) Obtain the mission action plans of the entities. Based on the problems and objectives of the simulation study and the operational action plan, clarify the mission types and mission sequences of each entity in the scenario.

[0185] (2) Planning the task sequence of entities. Based on the action plans of each entity during the operation, and according to the order and conditions of the tasks to be performed by each entity, the task sequence of each entity that needs to perform the planned tasks is set, and finally multiple sets of related action plans are formed.

[0186] (3) Run the simulation and monitor the execution status of entity tasks. Start the simulation, and according to the set action plan of the entities, the task scheduler will cyclically check the tasks that meet the execution conditions and schedule the entity tasks for execution. During the simulation, the execution status of each task of all entities will be updated in real time.

[0187] (4) Adjusting tasks. During the simulation, by observing the battlefield situation and task sequence status updates, the currently executing task can be terminated, the unexecuted tasks of any entity can be canceled and adjusted, and tasks can be dynamically added to entities in the future.

[0188] (5) Simulation End. The simulation ends when all entities complete their action plans or meet the simulation end conditions. The execution status of all tasks set before the simulation can be statistically analyzed and viewed.

[0189] Example 3

[0190] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a device for constructing a simulation entity task model, as disclosed in an embodiment of the present invention. Figure 6 The described apparatus for constructing the simulated entity task model is applied in the field of entity task simulation technology, and the embodiments of the present invention are not limited thereto. Figure 6 As shown, the apparatus for constructing the simulation entity task model may include the following operations:

[0191] S301, Entity Object Acquisition Module, is used to acquire n entity objects for which the task needs to be performed, where n is a positive integer;

[0192] S302, Simulation entity task model construction module, used to process the n entity objects to obtain the simulation entity task model;

[0193] S303, Task Action Plan Acquisition Module, used to acquire the task action plan of an entity;

[0194] S304, Task Action Plan Execution Module, used to process the task action plan of the entity using the simulated entity task model, and obtain the task action plan execution result;

[0195] S305, Evaluation module, used to evaluate the execution results of the task action plan and obtain the task action plan evaluation results.

[0196] Example 4

[0197] Please see Figure 7 , Figure 7 This is a schematic diagram of another device for constructing a simulation entity task model disclosed in an embodiment of the present invention. Figure 7 The described apparatus for constructing the simulated entity task model is applied in the field of entity task simulation technology, and the embodiments of the present invention are not limited thereto. Figure 7 As shown, the apparatus for constructing the simulation entity task model may include the following operations:

[0198] Memory 401 storing executable program code;

[0199] Processor 402 coupled to memory 401;

[0200] The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the method for constructing the simulation entity task model described in Embodiment 1 and Embodiment 2.

[0201] Example 5

[0202] This invention discloses a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program enables a computer to perform the steps in the method for constructing a simulated entity task model described in Embodiments 1 and 2.

[0203] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0204] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0205] Finally, it should be noted that the method and apparatus for constructing a simulation entity task model disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A construction method of a simulation entity task model, characterized by, The method comprises: S1, acquiring n entity objects needing to perform tasks, n being a positive integer; S2, processing the n entity objects to obtain a simulation entity task model, comprising: S21, processing the n entity objects to obtain n single entity object task models, comprising: S211, processing the n entity objects to obtain task types of the n entity objects; S212, processing the task types of the n entity objects by using a preset task manager to obtain task queues of the n entity objects; The task queue expression of the n entity objects is: wherein QNum is the total number of all entity task queues in a single simulation scenario, n is the total number of entities in the scenario, m is the number of task types of the i-th entity, T ij is the j-th task type of the i-th entity. S213, processing the task queues of the n entity objects to obtain n single entity object task models, comprising: S2131, using the task manager to perform task arrangement on the task types of the n entity objects to obtain task planning information; S2132, performing task execution monitoring on the task planning information to obtain execution state information; S2133, performing task adjustment on the execution state information to obtain n single entity object task models; S22, processing the n single entity object task models to obtain a simulation entity task model; S3, acquiring a task action plan of an entity; S4, processing the task action plan of the entity by using the simulation entity task model to obtain a task action plan execution result; S5, evaluating the task action plan execution result to obtain a task action plan evaluation result, comprising: The task action plan execution result is analyzed to obtain a task action plan object set A = {A1, A2, …, A n} and B = {B1, B2, …, B m}, the two object sets respectively represent the self and the other party, n is the number of self objects, m is the number of other objects, an index set X = {X1, X2, …, X d} is set for each object, d is the number of indexes, and an original data matrix R = (r ij ) l·d , wherein r ij represents the data size of the jth index of the ith object, and l = n + m. For each indicator X i = {r'X ij}, i = 1, 2,..., d, j = 1, 2,..., l have: E = XW e + b e wherein W e ,b e are the weight parameters and bias parameters of the linear embedding layer of the self-attention mechanism of the Transformer, respectively, and are obtained by training. adding a learnable position code to distinguish the position Z of each object: Z=E+P where P = {p i} is the position encoding of the object, p i = W e x i + b e , x i is the actual position of object i; calculating a first matrix Q: Q = ZW Q a second matrix K: K = ZW K a third matrix V: V = ZW V Where Q is used to determine the information needs of the current agent, K is used to judge the key descriptions of all agents, V is used to pass the information of itself to the previous agent, so as to judge the possible changes of the situation, and W Q K V are training matrices obtained by learning.​​ entering an attention mechanism to calculate an attention score A: where softmax is a softmax function, T is transpose, d k is a constant, is a scaling factor, which is used to avoid gradient vanishing and prevent the dot product from being too large. calculating an attention weighted result O: O=Attention(Q,K,V) wherein Attention represents the attention mechanism; inputting a Transformer encoder TramsformerEncoder: H=TramsformerEncoder(O) get H, H each row h of H i both incorporate the object i's sensitivity to the surrounding environment; performing task action plan execution scoring: wherein, W o ,b o are the weight parameters and bias parameters of the Transformer fully connected layer, respectively, both of which are obtained by training learning, and the task action plan execution score s i of the object i is obtained, i.e., the task action plan execution result.

2. The method of claim 1, wherein, The simulation entity task model comprises an entity task configuration subsystem, a task list maintenance subsystem, a task running scheduling subsystem, a task state monitoring subsystem and a task dynamic adjustment subsystem; The entity task configuration subsystem is in data connection with the task list maintenance subsystem, and is used for setting a single condition or a condition group for a single task item; The task list maintenance subsystem is in data connection with the task running scheduling subsystem, and is used for providing a man-machine interactive operation entrance for combat task planning; The task running scheduling subsystem is in data connection with the task state monitoring subsystem and the task dynamic adjustment subsystem, and is used for providing a secondary development interface specification of a task model; The task state monitoring subsystem is used for providing a task execution state monitoring function based on a task list; The task dynamic adjustment subsystem is used for providing a man-machine interactive function of adjusting a task execution strategy in real time.

3. The method of claim 1, wherein, The task action plan of the entity is processed by using the simulation entity task model, and a task action plan execution result is obtained, including: S41, obtaining a task action plan of an entity; S42, processing the task action plan of the entity by using a task configuration subsystem to obtain a task type and a task sequence of each entity in a scenario; S43, starting simulation of the task type and the task sequence of each entity in the scenario by using a task list maintenance subsystem to obtain task action plan information; S44, monitoring the task action plan information by using a task state monitoring subsystem, updating an execution state of each entity according to a monitoring result, and obtaining entity update information; S45, updating the task action plan information according to the entity update information by using a task dynamic adjustment subsystem to obtain a task action plan execution result.

4. A construction device of a simulation entity task model, applied to the construction method of the simulation entity task model according to any one of claims 1-3, characterized in that, The device comprises: An entity object acquisition module is configured to acquire n entity objects that need to perform tasks, n being a positive integer; A simulation entity task model construction module is configured to process the n entity objects to obtain a simulation entity task model; A task action plan acquisition module is configured to acquire a task action plan of an entity; A task action plan execution module is configured to process the task action plan of the entity by using the simulation entity task model to obtain a task action plan execution result; An evaluation module is configured to evaluate the task action plan execution result to obtain a task action plan evaluation result.

5. A construction apparatus of a simulation entity task model, characterized by, The device comprises: A memory storing executable program codes; A processor coupled with the memory; The processor invokes the executable program codes stored in the memory to execute the simulation entity task model construction method according to any one of claims 1-3.

6. A computer storable medium, characterized by The computer storage medium stores computer instructions, which are invoked to execute the simulation entity task model construction method according to any one of claims 1-3.

Images