An agent-based simulation method and system

Abstract

The disclosure provides a simulation method and system based on agent technology, and belongs to the technical field of simulation. The method comprises the following steps: in response to receiving a simulation instruction, starting a first process for initialization; starting at least one agent process, at least one second process and at least one third process; the first process, the at least one agent process, the at least one second process and the at least one third process are respectively in a subscription relationship with the related message topics in the at least one message topic and publish the running results on the corresponding message topics, thereby generating a simulation action. Compared with the prior art, the method and system provided by the disclosure integrate the agent technology, the subscription-published mechanism and the multi-process mechanism, provide a simulation scheme with low coupling and self-feedback decision force, and thus realize a more rapid and accurate simulation result.

Description

Technical Field

[0001] This invention belongs to the field of simulation technology, and in particular relates to a simulation method and system based on intelligent agent technology. Background Technology

[0002] Simulation technology is now widely used in various industries, such as manufacturing and gaming. It allows for the simulation of changing real-world environments, enabling rapid business iteration and reducing unnecessary expenses. However, existing simulation programs generally suffer from limitations in execution speed, scalability, and a lack of robust decision-making mechanisms, resulting in simulation results that fail to fully meet business needs. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention proposes a simulation method and system based on intelligent agent technology.

[0004] The first aspect of this invention proposes a simulation method based on intelligent agent technology, the method comprising: In response to receiving a simulation command, the first process is started for initialization; wherein, initialization includes simulation initialization and communication initialization, and communication initialization includes creating at least one message topic based on the task type of the simulation command; Start at least one agent process, at least one second process, and at least one third process; The first process, at least one agent process, at least one second process, and at least one third process shall each establish a subscription relationship with a relevant message topic in at least one message topic; The first subscription data is obtained from the relevant message topics subscribed to using at least one intelligent agent process; Based on the first subscription data, at least one intelligent agent process is used to generate decision data and publish the decision data to the relevant message topics subscribed to by the first process. Utilize the first process to obtain decision data from subscribed relevant message topics; Based on decision data, at least one real-time computing subtask data is generated by the first process and published to at least one relevant message topic subscribed to by the second process. At least one second process is used to obtain at least one real-time computing subtask data from subscribed relevant message topics; Based on data from at least one real-time computing subtask, at least one behavioral data is generated using at least one second process and published to a relevant message topic subscribed to by at least one third process. Use at least one third process to obtain at least one behavioral data point from the relevant message topics subscribed to; Based on at least one behavioral data, corresponding simulation actions are generated using the at least one third process.

[0005] A second aspect of the present invention provides a simulation system based on intelligent agent technology, the system comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to implement the method as described in the first aspect of the present invention.

[0006] In summary, the simulation method and system proposed in this invention integrates agent technology, a publish-subscribe mechanism, and a parallel operation mechanism. By utilizing agent technology and a self-motivating mechanism, the simulation decision-making capability is improved. At the same time, the publish-subscribe mechanism reduces program coupling, and the multi-process mechanism enhances parallel operation capability, thereby significantly improving simulation efficiency and result accuracy. Attached Figure Description

[0007] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0008] Figure 1 This is a flowchart illustrating a simulation method based on intelligent agent technology proposed in an embodiment of the present invention.

[0009] Figure 2 This is a schematic diagram of the structure of a simulation system based on intelligent agent technology proposed in an embodiment of the present invention. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0011] Appendix Figure 1 The first aspect of the present invention illustrates a simulation method based on intelligent agent technology, the method comprising: Step 101: In response to receiving the simulation command, the first process is initiated for initialization. Initialization includes simulation initialization and communication initialization. Communication initialization includes creating at least one message topic based on the task type of the simulation command. Simulation initialization includes configuring simulation target parameters and simulation environment parameters. For example, in a simulated air combat game, simulation initialization includes setting weather conditions, temperature conditions, number of aircraft, number of firepower, combat rules, and geographical maps. Simultaneously, the system's monitoring program can be initiated to monitor the simulation frame rate and resource usage. Understandably, after initialization, the first process publishes necessary information on relevant message topics for subsequent processes subscribed to those topics to read and execute subsequent simulation processing.

[0012] In one alternative embodiment, necessary agent processes can be added to the already running agent processes. For example, in a simulated air combat game, air combat agents and route planning agents exist in advance, and manned aircraft agents and unmanned aerial vehicle agents can be added to make more comprehensive simulation decisions.

[0013] In one optional embodiment, adding a required intelligent agent process to an already running intelligent agent process includes, in response to receiving a packaged functional model unit, parsing the functional model unit to obtain the relevant files for the new intelligent agent; running the relevant files, and adding a new intelligent agent process to at least one intelligent agent process. In one optional embodiment, the relevant files may include source code files, model description files, and information description files. The source code files can be configured as dynamic link libraries that can be loaded on different operating system platforms. The model description files describe the basic parameter information of the intelligent agent, and the information description files store descriptive information about the functional model units, which helps in understanding the basic information of the intelligent agent. The model description files are used by the main process to read the intelligent agent's information to configure the intelligent agent's input / output ports, variables, structure, logs, basic operating parameters, etc., thereby enabling the intelligent agent process to start smoothly and communicate effectively.

[0014] Step 102: Start at least one agent process, at least one second process, and at least one third process. It is understood that the at least one agent process, at least one second process, and at least one third process are typically started by the already running first process.

[0015] Step 103: Establish subscription relationships between the first process, at least one agent process, at least one second process, and at least one third process and the relevant message topics in at least one message topic.

[0016] Typically, processes of different types subscribe to different topics, although some overlap is possible; while processes of the same type subscribe to the same topics. However, this embodiment of the invention does not limit this, and processes of different types can also subscribe to the same topics, for example, the first process and the second process can subscribe to all topics.

[0017] For example, in a simulated air combat game, a route planning agent can subscribe to terrain data, aircraft status information, and battlefield situational messages, while the first process can subscribe to combat command messages. Understandably, once a process subscribes to a message topic, newly published information on that topic can be obtained by the process.

[0018] Step 104: Use at least one agent process to retrieve first subscription data from the relevant subscribed message topics. In an alternative embodiment, multiple agent processes may communicate to exchange the retrieved data.

[0019] Step 105: Based on the first subscribed data, generate decision data using at least one agent process and publish the decision data to the relevant message topics subscribed to by the first process. For example, in a simulated air combat game, the air combat agent can generate combat commands for aircraft movement and fire control as decision data.

[0020] Step 106: Use the first process to obtain decision data from the relevant message topics you have subscribed to.

[0021] Step 107: Based on the decision data, use the first process to generate at least one real-time computational subtask data and publish it to at least one relevant message topic subscribed to by the second process. For example, in a simulated air combat game, the decision data of the combat command is broken down into subtasks such as calculating the aircraft's movement speed, calculating the aircraft's movement time, and calculating the aircraft's firing time.

[0022] Step 108: Utilize at least one second process to obtain at least one real-time computing subtask data from subscribed relevant message topics. In an optional embodiment, multiple second processes may communicate to exchange the obtained data.

[0023] Step 109: Based on data from at least one real-time computing subtask, generate at least one set of behavioral data using at least one second process and publish it to a relevant message topic subscribed to by at least one third process. For example, in a simulated air combat game, the behavioral data could be the specific flight speed and time of each aircraft, as well as the specific firing time of each aircraft.

[0024] In an optional embodiment, generating at least one behavioral data using at least one second process includes generating at least one behavioral data using an idle process among the at least one second process. It is understood that the multiple second processes can communicate to assign tasks to idle second processes, thereby ensuring the speed of the simulation execution.

[0025] Step 110: Utilize at least one third process to obtain at least one behavioral data point from the subscribed relevant message topics. In an optional embodiment, multiple third processes may communicate to exchange the obtained data.

[0026] Step 111: Based on at least one behavioral data point, generate corresponding simulation actions using at least one third process. For example, in a simulated air combat game, the third process can render aircraft performing flight and firing actions, etc.

[0027] In an optional embodiment, at least one third process can be used to collect real-time simulation state data after generating the corresponding simulation actions; at least one third process can also be used to publish the simulation state data to relevant message topics subscribed to by at least one agent. For example, in a simulated air combat game, after an aircraft flies and fires, it is damaged, and the terrain and the number of aircraft also change. At this time, the third process collects this real-time simulation state data and publishes it through message topics.

[0028] In an optional embodiment, at least one agent process can be used to obtain real-time simulation state data from relevant message topics; based on the real-time simulation state data, at least one agent process can generate new decision data and publish the new decision data to the relevant message topics subscribed to by the first process. In this way, the agents can continuously make new dynamic decisions based on the progress of the simulation, thereby forming more reasonable simulation results.

[0029] In an optional embodiment, at least one agent process can be used to extract relevant state data from the simulated real-time state data; the state data can be compared with a preset self-excitation threshold using at least one agent process; and self-excitation can be applied to at least one agent process based on the comparison result. For example, for a path planning agent, the self-excitation threshold is that each decision allows the simulated vehicle to travel 20 kilometers in 10 minutes. However, the relevant state data shows that the simulated vehicle only traveled 15 kilometers in 10 minutes after this decision, which forms a negative incentive that affects the path planning agent's decision adjustment. It is understood that positive incentives can also be formed to influence the decision, and the change in state can be determined by the extracted relevant state data or by comparing the extracted relevant state data with the previously extracted relevant state data.

[0030] In an optional embodiment, at least one third process can be used to collect simulation real-time status data, and then at least one third process can be used to compare the simulation real-time status data with a preset termination condition; if the preset termination condition is met, at least one third process can send a simulation termination command to the first process.

[0031] Figure 2 The structure of a simulation system 200 based on intelligent agent technology according to a second aspect of the present invention is shown. The system 200 includes: at least one processor 201; and a memory 202 communicatively connected to the at least one processor; wherein the memory 202 stores instructions executable by the at least one processor 201, the instructions being executed by the at least one processor 201 to enable the at least one processor 201 to implement the following method steps: Step 1: In response to receiving the simulation command, the first process is initiated for initialization. Initialization includes simulation initialization and communication initialization. Communication initialization involves creating at least one message topic based on the task type of the simulation command. Simulation initialization includes configuring simulation target parameters and simulation environment parameters. For example, in a simulated air combat game, simulation initialization includes setting weather conditions, temperature conditions, number of aircraft, number of firepower, combat rules, and geographical maps. It can also initiate a system monitoring program to monitor the simulation frame rate and resource usage. Understandably, after initialization, the first process publishes necessary information to the relevant message topics for subsequent processes subscribed to those topics to read and execute subsequent simulation processing.

[0032] In one alternative embodiment, necessary agent processes can be added to the already running agent processes. For example, in a simulated air combat game, air combat agents and route planning agents exist in advance, and manned aircraft agents and unmanned aerial vehicle agents can be added to make more comprehensive simulation decisions.

[0033] In one optional embodiment, adding a required intelligent agent process to an already running intelligent agent process includes, in response to receiving a packaged functional model unit, parsing the functional model unit to obtain the relevant files for the new intelligent agent; running the relevant files, and adding a new intelligent agent process to at least one intelligent agent process. In one optional embodiment, the relevant files may include source code files, model description files, and information description files. The source code files can be configured as dynamic link libraries that can be loaded on different operating system platforms. The model description files describe the basic parameter information of the intelligent agent, and the information description files store descriptive information about the functional model units, which helps in understanding the basic information of the intelligent agent. The model description files are used by the main process to read the intelligent agent's information to configure the intelligent agent's input / output ports, variables, structure, logs, basic operating parameters, etc., thereby enabling the intelligent agent process to start smoothly and communicate effectively.

[0034] Step 2: Start at least one agent process, at least one second process, and at least one third process. It is understood that the at least one agent process, at least one second process, and at least one third process are typically started by the already running first process.

[0035] Step 3: Establish subscription relationships between the first process, at least one agent process, at least one second process, and at least one third process and the relevant message topics in at least one message topic.

[0036] Typically, processes of different types subscribe to different topics, although some overlap is possible; while processes of the same type subscribe to the same topics. However, this embodiment of the invention does not limit this, and processes of different types can also subscribe to the same topics, for example, the first process and the second process can subscribe to all topics.

[0037] For example, in a simulated air combat game, a route planning agent can subscribe to terrain data, aircraft status information, and battlefield situational messages, while the first process can subscribe to combat command messages. Understandably, once a process subscribes to a message topic, newly published information on that topic can be obtained by the process.

[0038] Step 4: Utilize at least one agent process to retrieve first-subscribed data from the relevant message topics. In an optional embodiment, multiple agent processes may communicate to exchange the retrieved data.

[0039] Step 5: Based on the first subscribed data, generate decision data using at least one agent process and publish the decision data to the relevant message topics subscribed to by the first process. For example, in a simulated air combat game, the air combat agent can generate combat commands for aircraft movement and fire control as decision data.

[0040] Step Six: Use the first process to obtain decision data from the relevant message topics you have subscribed to.

[0041] Step 7: Based on the decision data, use the first process to generate at least one real-time computational subtask data and publish it to at least one relevant message topic subscribed to by the second process. For example, in a simulated air combat game, the decision data for combat commands is broken down into subtasks such as calculating the aircraft's movement speed, calculating the aircraft's movement time, and calculating the aircraft's firing time.

[0042] Step 8: Utilize at least one second process to retrieve at least one real-time computing subtask data from the subscribed relevant message topics. In an optional embodiment, multiple second processes may communicate to exchange the retrieved data.

[0043] Step Nine: Based on data from at least one real-time computing subtask, generate at least one set of behavioral data using at least one second process and publish it to a relevant message topic subscribed to by at least one third process. For example, in a simulated air combat game, behavioral data could be the specific flight speed and time of each aircraft, as well as the specific firing time of each aircraft.

[0044] In an optional embodiment, generating at least one behavioral data using at least one second process includes generating at least one behavioral data using an idle process among the at least one second process. It is understood that the multiple second processes can communicate to assign tasks to idle second processes, thereby ensuring the speed of the simulation execution.

[0045] Step 10: Utilize at least one third process to retrieve at least one behavioral data point from the subscribed relevant message topics. In an optional embodiment, multiple third processes may communicate to exchange the retrieved data.

[0046] Step 11: Based on at least one behavioral data point, generate corresponding simulation actions using at least one third process. For example, in a simulated air combat game, the third process can render aircraft performing flight and firing actions.

[0047] In an optional embodiment, at least one third process can be used to collect real-time simulation state data after generating the corresponding simulation actions; at least one third process can also be used to publish the simulation state data to relevant message topics subscribed to by at least one agent. For example, in a simulated air combat game, after an aircraft flies and fires, it is damaged, and the terrain and the number of aircraft also change. At this time, the third process collects this real-time simulation state data and publishes it through message topics.

[0048] In an optional embodiment, at least one agent process can be used to obtain real-time simulation state data from relevant message topics; based on the real-time simulation state data, at least one agent process can generate new decision data and publish the new decision data to the relevant message topics subscribed to by the first process. In this way, the agents can continuously make new dynamic decisions based on the progress of the simulation, thereby forming more reasonable simulation results.

[0049] In an optional embodiment, at least one agent process can be used to extract relevant state data from the simulated real-time state data; the state data can be compared with a preset self-excitation threshold using at least one agent process; and self-excitation can be applied to at least one agent process based on the comparison result. For example, for a path planning agent, the self-excitation threshold is that each decision allows the simulated vehicle to travel 20 kilometers in 10 minutes. However, the relevant state data shows that the simulated vehicle only traveled 15 kilometers in 10 minutes after this decision, which forms a negative incentive that affects the path planning agent's decision adjustment. It is understood that positive incentives can also be formed to influence the decision, and the change in state can be determined by the extracted relevant state data or by comparing the extracted relevant state data with the previously extracted relevant state data.

[0050] In an optional embodiment, at least one third process can be used to collect simulation real-time status data, and then at least one third process can be used to compare the simulation real-time status data with a preset termination condition; if the preset termination condition is met, at least one third process can send a simulation termination command to the first process.

[0051] In summary, the simulation method and system proposed in this invention integrates agent technology, a publish-subscribe mechanism, and a parallel operation mechanism. By utilizing agent technology and a self-motivating mechanism, the simulation decision-making capability is improved. At the same time, the publish-subscribe mechanism reduces program coupling, and the multi-process mechanism enhances parallel operation capability, thereby significantly improving simulation efficiency and result accuracy.

[0052] It should be noted that, in the several embodiments provided by this invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units or steps is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or steps may be combined or integrated into another system or step, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0053] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0054] Furthermore, in the various embodiments of the present invention, the functional units or method steps can be integrated into one processing unit or one method step, or each unit or method step can exist separately physically or functionally, or two or more units can be integrated into one unit. The integrated unit described above can be implemented in hardware or in the form of hardware plus software functional units.

[0055] It should be noted that the embodiments provided in this invention are all illustrative, and different embodiments can be arbitrarily and reasonably combined. For the sake of brevity, not all possible combinations of the various technical features in the above embodiments are described; however, as long as such combinations do not contradict each other, they should all be considered to fall within the scope of this specification. Furthermore, it should be understood that the systems and methods disclosed in the embodiments provided in this invention can be implemented in other ways or with modifications. Any substitutions made in hardware or software, or any modifications made without departing from the concept of this invention, are within the protection scope of this invention.

Claims

1. A simulation method based on intelligent agent technology, characterized in that, include: In response to receiving a simulation command, a first process is initiated for initialization; wherein, the initialization includes simulation initialization and communication initialization, and the communication initialization includes creating at least one message topic based on the task type of the simulation command; Start at least one agent process, at least one second process, and at least one third process; The first process, the at least one agent process, the at least one second process, and the at least one third process shall respectively establish subscription relationships with the relevant message topics in the at least one message topic; The first subscription data is obtained from the subscribed related message topics using the at least one intelligent agent process; Based on the first subscription data, decision data is generated using the at least one intelligent agent process and the decision data is published in the relevant message topic subscribed to by the first process; The decision data is obtained from the subscribed relevant message topics using the first process; Based on the decision data, the first process generates at least one real-time computing subtask data and publishes it to the relevant message topic subscribed to by the at least one second process; The at least one second process is used to obtain the at least one real-time computing subtask data from the subscribed relevant message topics; Based on the at least one real-time computing subtask data, at least one behavioral data is generated using the at least one second process and published to the relevant message topic subscribed to by the at least one third process; The at least one third process is used to obtain the at least one behavioral data from the subscribed relevant message topics; Based on the at least one behavioral data, corresponding simulation actions are generated using the at least one third process.

2. The simulation method based on intelligent agent technology according to claim 1, characterized in that: In response to receiving the encapsulated functional model unit, the functional model unit is parsed to obtain the relevant files of the new intelligent agent; Run the relevant files to add a new agent process to the at least one agent process.

3. The simulation method based on intelligent agent technology according to claim 2, characterized in that... : The relevant files include source code files, model description files, and information description files.

4. The simulation method based on intelligent agent technology according to claim 1, characterized in that: The at least one third process is used to collect real-time simulation status data after generating the corresponding simulation actions; The simulation state data is published to the relevant message topics subscribed to by the at least one intelligent agent using the at least one third process.

5. The simulation method based on intelligent agent technology according to claim 4, characterized in that: The simulation real-time state data is obtained from the relevant message topics using at least one intelligent agent process; Based on the simulated real-time state data, new decision data is generated using the at least one intelligent agent process, and the new decision data is published to the relevant message topics subscribed to by the first process.

6. The simulation method based on intelligent agent technology according to claim 5, characterized in that: The at least one intelligent agent process is used to extract relevant state data from the simulation real-time state data; The state data is compared with a preset self-excitation threshold using the at least one intelligent agent process; Based on the results of the comparison, the at least one agent process is self-energized.

7. The simulation method based on intelligent agent technology according to claim 4, characterized in that: After collecting the simulation real-time status data using at least one third process, the simulation real-time status data is compared with a preset termination condition using at least one third process. If the preset termination condition is met, the at least one third process sends a simulation termination command to the first process.

8. The simulation method based on intelligent agent technology according to claim 1, characterized in that, The generation of at least one behavioral data using the at least one second process includes generating the at least one behavioral data using an idle process among the at least one second process.

9. A simulation method based on intelligent agent technology according to claim 1, characterized in that, The simulation initialization includes configuring the simulation target parameters and simulation environment parameters.

10. A simulation system based on intelligent agent technology, the system comprising: At least one processor; And a memory communicatively connected to at least one processor; wherein the memory stores instructions executable by at least one processor, the instructions being executed by at least one processor to enable the at least one processor to implement the method as described in any one of claims 1-9.

Images