Multi-agent application system implementation processing method and system, intelligent terminal and medium

Abstract

The invention discloses a multi-agent application system implementation processing method and system, an intelligent terminal and a medium, and the method comprises the steps: presetting a multi-agent system, and setting cooperation modes of a cooperation mode, a competition mode, a mixed mode and a profit mode among agents supported by the multi-agent system, the method is used for adapting to different application scenes and task requirements. When the multi-agent system receives multiple tasks, the currently received multiple tasks are analyzed; judging a cooperation mode required by the currently received multiple tasks; and controlling the multi-agent system to adopt a corresponding cooperation mode to execute and complete the currently received multiple tasks. When the multi-agent application system is used for processing complex tasks, the tasks can be effectively decomposed, resources can be effectively distributed, decisions can be effectively optimized, and convenience is provided for use of users.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a multi-agent application system implementation processing method, system, intelligent terminal and storage medium. Background Art

[0002] With the development of artificial intelligence technology, multi-agent systems have become a key technology for solving complex problems. Traditional single-agent systems have limitations in handling complex tasks and cannot effectively decompose tasks, allocate resources, and optimize decisions.

[0003] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a multi-agent application system implementation processing method, system, intelligent terminal and storage medium in response to the problems and defects of the above-mentioned prior art. The purpose of the present invention is to provide an application framework for a multi-agent system, which can support collaboration, competition and hybrid modes among agents to adapt to different application scenarios and task requirements, and provide convenience for users.

[0005] The technical solution adopted by the present invention to solve the problem is as follows: A multi-agent application system implementation processing method, comprising: Pre-setting a multi-agent system, and setting the multi-agent system to support cooperation modes among agents, such as cooperation mode, competition mode, hybrid mode and egoistic mode, so as to adapt to different application scenarios and task requirements; When the multi-agent system receives multiple tasks, it analyzes the currently received multiple tasks and determines the cooperation mode required by the currently received multiple tasks; When it is determined that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-tasks is the cooperative mode; When it is determined that the currently received multi-tasks require agents to compete for limited resources or pursue individual interests maximization, the cooperation mode required by the currently received multi-tasks is determined to be a competition mode; When it is determined that the currently received multi-task requires the agents to cooperate in some cases and to compete in other cases, it is determined that the cooperation mode required by the currently received multi-task is a mixed mode; When it is judged that the currently received multi-task requires the agent to give priority to and achieve its own interests or goals, the self-interest mode required by the currently received multi-task is determined; Control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks.

[0006] The multi-agent application system implementation processing method, wherein the step of controlling the multi-agent system to adopt a corresponding collaboration mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the cooperative mode, the control agents work together to achieve a common goal or optimize the utilization of a shared resource. The execution steps include: Control effective communication and information sharing between agents, and share each other's status, intentions, and action plans; Control each agent to make autonomous decisions when faced with problems, and formulate appropriate cooperation strategies to allocate resources, divide tasks, and resolve conflicts; Control each intelligent agent to coordinate their actions based on shared information and cooperation strategies to achieve common goals.

[0007] The multi-agent application system implementation processing method, wherein the step of controlling the multi-agent system to adopt a corresponding collaboration mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the competition mode, the control agents compete for limited resources or pursue individual interests maximization. The execution steps include: Control each agent to learn the opponent's strategy and behavior pattern, and model the opponent to predict its behavior and decision; Control each agent to set a corresponding competitive strategy to pursue individual interests, including selecting optimized actions, adjusting strategies, and predicting the opponent's behavior; Control each intelligent agent to achieve better results in the competition for resources by optimizing competition strategies.

[0008] The multi-agent application system implementation processing method, wherein the step of controlling the multi-agent system to adopt a corresponding collaboration mode to execute and complete the currently received multiple tasks includes: When the corresponding cooperation mode is a mixed mode, the control agents need to cooperate in some cases and show competitive characteristics in other cases. The execution steps include: Each agent is equipped with situational awareness to identify when cooperation and competition are needed; Control each intelligent agent to dynamically adjust cooperation or competition strategies according to environmental changes and task requirements; Control each intelligent agent to dynamically collaborate and compete according to actual conditions during the execution of tasks, so as to adapt to the ever-changing environment and goals.

[0009] The multi-agent application system implementation processing method, wherein the step of controlling the multi-agent system to adopt a corresponding collaboration mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is the self-interest mode, the control agent prioritizes and realizes its own interests or goals. The execution steps include: Control each agent to optimize its individual goals when performing tasks and interacting with other agents; Control each intelligent agent to optimize resource allocation according to resource utilization; Control each intelligent agent to make independent decisions based on its own needs. The multi-agent application system implements the processing method, wherein the agents include: robots, drones, cameras, traffic lights and / or medical equipment.

[0010] The multi-agent application system implementation processing method, wherein the step of controlling the multi-agent system to adopt a corresponding collaboration mode to execute and complete the currently received multiple tasks also includes: When each agent detects task priority adjustments and changes in resource availability during the task, the control re-evaluates and switches to a collaborative mode that matches the changed tasks and resources.

[0011] A multi-agent application system implementation processing system, wherein the system comprises: A presetting module is used to pre-set a multi-agent system, and set the multi-agent system to support a cooperation mode, a competition mode, a mixed mode and a self-interest mode among the agents, so as to adapt to different application scenarios and task requirements; A collaboration mode analysis and judgment module is used to analyze the currently received multi-tasks when the multi-agent system receives multiple tasks; to judge the collaboration mode required by the currently received multi-tasks; when it is judged that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-tasks is a collaboration mode; when it is judged that the currently received multi-tasks require the agents to compete for limited resources or pursue the maximization of individual interests, it is determined that the collaboration mode required by the currently received multi-tasks is a competition mode; when it is judged that the currently received multi-tasks require the agents to cooperate in some cases and show competitive characteristics in other cases, it is determined that the collaboration mode required by the currently received multi-tasks is a mixed mode; when it is judged that the currently received multi-tasks require the agents to give priority to and achieve their own interests or goals, it is determined that the currently received multi-tasks require a selfish mode; The task execution control module is used to control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks.

[0012] An intelligent terminal includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, including the method for executing any one of the methods described above.

[0013] A computer-readable storage medium, wherein when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any one of the methods described above.

[0014] Beneficial effects of the invention: The invention provides a method, system, intelligent terminal and storage medium for realizing a multi-agent application system. The multi-agent system provided by the invention uses distributed artificial intelligence technology to decompose a complex system into multiple small, mutually serving system units, and achieves the overall goal of the system through mutual communication, sharing of information and resources, and collaboration and competition. The invention includes autonomy, local perspective, decentralization, collaboration, interactive communication, and can adapt to collaborative modes of cooperation, competition, mixed mode and egoistic mode; the multi-agent application system of the invention can effectively decompose tasks, allocate resources and optimize decisions when processing complex tasks, providing convenience for users. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of the multi-agent application system implementation processing method provided in Example 1 of the present invention.

[0017] Figure 2 It is a flow chart of the multi-agent application system implementation processing method provided in Example 2 of the present invention.

[0018] Figure 3 A principle block diagram of an embodiment of a multi-agent application system processing system provided by the present invention.

[0019] Figure 4 It is a block diagram of the internal structure principle of the intelligent terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention clearer and more specific, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0021] Single-agent systems based on existing technologies have limitations when dealing with complex tasks and are unable to effectively decompose tasks, allocate resources, and optimize decision-making.

[0022] The present invention provides an application framework for a multi-agent system, which can support collaboration, competition and hybrid modes among agents to adapt to different application scenarios and task requirements. The key features of the present invention include autonomy, local perspective, decentralization, collaboration, interactive communication, and can adapt to collaborative modes of cooperation, competition, hybrid mode and egoistic mode.

[0023] The Multi-Agent System (MAS) is a system composed of multiple agents that can accomplish specific tasks or goals through mutual collaboration, communication, and competition. Agents can be autonomous software programs, robots, or other entities that can perceive the environment and take actions.

[0024] The application framework of a multi-agent system in the embodiment of the present invention emphasizes its flexibility and adaptability in different task requirements. The key features are analyzed as follows: Autonomy: This means that the agent of the present invention can make decisions and perform tasks independently without relying on central control. This autonomy makes the system more flexible and responsive.

[0025] Having a local perspective means that each agent in the embodiment of the present invention only obtains and processes a part of the information of its surrounding environment; this can reduce the complexity of information processing and improve the efficiency of the system.

[0026] Decentralization: This means that the control and decision-making of the present invention are decentralized among multiple intelligent agents, avoiding the risk of single point failure and correspondingly improving the robustness and reliability of the system.

[0027] Collaborative: refers to the fact that the intelligent agents of the present invention can complete tasks together through cooperation, such as sharing the workload, which can improve the efficiency and effectiveness of task completion.

[0028] Interactive communication: refers to the fact that the intelligent agents of the present invention enhance the effectiveness and efficiency of cooperation by exchanging information with each other. Effective communication enables the intelligent agents to adjust their respective strategies in real time.

[0029] The present invention can also adapt to a variety of collaboration modes: supporting cooperation, competition, mixed mode and egoistic mode, and can flexibly adjust its collaboration strategy in different environments and task requirements.

[0030] In specific applications, for example, the multi-agent application system of the present invention is applied in an intelligent traffic management system, and each traffic light can be used as an agent to autonomously adjust the signal time based on real-time traffic flow data through the method of the present invention. They can cooperate to reduce overall traffic congestion, and can also compete in certain circumstances to give priority to emergency vehicles.

[0031] For another example, the multi-agent application system of the present invention is applied to a drone group. In the drone delivery system, each drone can act as an independent agent and autonomously select the optimal path through the control method of the present invention. When performing tasks, they can share information, avoid encounters, optimize delivery time and reduce energy consumption.

[0032] For another example, the multi-agent application system of the present invention is applied to a multi-player online game. The computer-controlled characters (NPCs) controlled by the method of the present invention can make autonomous decisions, cooperate or compete according to the player's behavior to increase the fun and challenge of the game.

[0033] The following advantages can be achieved by the present invention: 1) Improved efficiency: Decentralized decision-making and collaboration among agents can speed up problem solving, especially when dealing with complex or large-scale tasks.

[0034] 2) Enhanced flexibility: Adjust the collaboration mode according to real-time information and environmental changes so that the system can quickly adapt to different situations.

[0035] 3) Improved robustness: The system can tolerate the failure of individual agents because the overall operation does not depend on a single control point.

[0036] 4) Optimizing resource utilization: Through collaboration between agents, resources can be used more efficiently, reducing redundancy and waste.

[0037] Specifically, Figure 1 As shown, a multi-agent application system implementation processing method of an embodiment of the present invention includes the following steps: Step S100, pre-setting a multi-agent system, and setting the multi-agent system to support cooperation mode, competition mode, hybrid mode and egoistic mode among agents, so as to adapt to different application scenarios and task requirements; In the embodiments of the present invention, the preset configuration of the multi-agent system (MAS) is firstly performed to enable it to flexibly adapt to different collaboration modes according to specific application scenarios and task requirements. These modes include cooperative mode, competitive mode, hybrid mode and egoistic mode. The following will explain each mode in detail, as well as their benefits and provide relevant application examples.

[0038] In the cooperative mode, the agents are configured to collaborate with each other to complete tasks together. The agents complement each other by sharing information and resources, thereby maximizing efficiency. For example, in an environmental monitoring system, multiple drones can share data to accurately monitor climate change in a large area.

[0039] In the competitive mode, each agent is configured to pursue its own goals and may conflict with other agents. In this mode, each agent relies on its own strategy to win resources or tasks. This mode configuration is used in auction or market simulations, for example, companies represented by multiple agents compete in auctions to obtain contracts or resources.

[0040] The hybrid mode configuration combines the characteristics of cooperation and competition and is suitable for complex application scenarios. In this mode, the control agents can cooperate in some scenarios and adopt a competitive strategy in other situations. For example, in a multi-robot rescue mission, the robots can cooperate to search for trapped people, but when rescue resources are limited, they may compete for resources.

[0041] In the egoistic mode, the agent is configured to focus primarily on its own interests, prioritizing the achievement of personal goals rather than collective goals. This mode is often used in situations where resources are limited. For example, in an online game, players may choose to focus only on their own score and survival without cooperating with other players.

[0042] Specific application configuration example: 1) In the intelligent transportation system, traffic lights and surveillance cameras are set as intelligent agents, which can switch to cooperative mode according to the current traffic conditions, share traffic information and adjust signal duration to reduce congestion. At the same time, the system also supports competitive mode, giving certain emergency vehicles priority to pass.

[0043] 2) In autonomous vehicle applications, when autonomous vehicles are set up to travel on highways, they can use cooperative mode to drive in a convoy, reduce air resistance and improve fuel efficiency. In urban environments, configurations may face competitive modes, such as competing for limited parking spaces or the right of way at intersections.

[0044] 3) In logistics and warehouse management, different handling robots can be set up to jointly manage the handling of goods using a cooperative mode and share inventory information in real time. During peak demand periods, these robots may switch to a competitive mode to prioritize orders to be processed.

[0045] It can be seen that the present invention can effectively improve the adaptability and performance of the multi-agent system by presetting different collaboration modes for the system, and meet the requirements of various application scenarios and tasks; it enables various types of agents to be reasonably deployed according to specific situations, thereby optimizing the performance of the overall system.

[0046] By presetting different collaboration modes, the present invention enables the multi-agent system to flexibly adjust its working mode according to environmental changes and task requirements, thereby improving overall performance. In the appropriate mode, the agents can optimize resource usage, reduce duplication of work, and improve the efficiency of collaboration and decision-making.

[0047] And for complex tasks that require the participation of multiple parties, by switching between different modes, the agents can effectively cope with various challenges, such as real-time coordination and strategy adjustment. And when one agent encounters a problem, other agents can adapt by changing the cooperation mode to ensure the stability and continuous operation of the system.

[0048] Step S200: When the multi-agent system receives multiple tasks, the multi-agent system analyzes the currently received multiple tasks and determines the cooperation mode required by the currently received multiple tasks. In the embodiment of the present invention, when the multi-agent system receives multiple tasks, it first analyzes these tasks and determines the required collaboration mode. Specifically, regarding task reception and analysis, when the multi-agent system receives multiple tasks from the environment or users, the system of the embodiment of the present invention will start the analysis process of these tasks. This step includes identifying factors such as the nature of the task, resource requirements, time constraints, and expected results. Through analysis, the present invention can clarify the characteristics of each task and the conditions required to complete these tasks.

[0049] Regarding the judgment of the collaboration mode, specifically, after analysis, the system of the present invention can judge the most suitable collaboration mode (such as cooperation, competition, mixed or selfish mode) according to the characteristics of the task. For example, if the task requires teamwork to achieve a shared goal, the present invention will judge to choose the cooperation mode; and if there is resource competition between tasks, the present invention will judge to adopt the competition mode. This judgment not only takes into account the needs of the task itself, but also comprehensively considers environmental factors and the capabilities of the intelligent agent.

[0050] For example, in a disaster rescue scenario, a multi-agent system receives multiple tasks, such as finding survivors, assessing damage, and transporting supplies. The embodiment of the present invention first analyzes these tasks, determines which tasks require robot collaboration (such as jointly searching an area) and which tasks may involve resource competition (such as transporting limited medical supplies), and then selects a suitable collaboration mode. In this way, rescue work can be organized and executed more efficiently, maximizing the success rate.

[0051] For another example, in the application scenario of a smart manufacturing factory, multiple intelligent robots receive multiple production tasks. These tasks may involve different production lines and resources. When the system analysis shows that some tasks require a high degree of collaboration (such as working together on the assembly line), while other tasks are independent choices, the corresponding robot group can be assigned to the collaborative tasks, while the independent tasks can be assigned to other robots. This dynamic judgment and on-demand adjustment can significantly improve production efficiency.

[0052] For another example, in the application scenario of intelligent traffic management, the image processing and data sensors can simultaneously transmit multiple traffic management tasks to the traffic signal control system, such as adjusting the traffic lights as an intelligent body, guiding traffic flow, and diverting accident scenes. When analysis finds that some tasks require mutual cooperation (such as coordinated adjustment of traffic lights), while other tasks are performed independently (such as synchronization of a single traffic light), different cooperation modes can be controlled to ensure smooth traffic and timely response to accidents.

[0053] It can be seen that by analyzing multiple tasks and determining the cooperation mode, the multi-agent system can optimize the action strategy in a variety of application scenarios. This not only improves the efficiency of task completion, but also provides more powerful support for solving complex problems.

[0054] By analyzing tasks reasonably and selecting appropriate collaboration modes, the present invention enables agents to collaborate more efficiently, reducing waste of time and resources and improving task execution efficiency. Accurate judgment of received multi-task requirements helps to reasonably allocate agent resources, ensure optimal resource use, and optimize resource allocation.

[0055] The present invention enables the multi-agent system to quickly adapt to changing task requirements and environmental conditions by dynamically analyzing and selecting different cooperation modes, thereby enhancing the flexibility of the system. It can also reduce conflicts and errors: by clarifying the cooperation mode, the agents can reduce conflicts and misunderstandings when performing tasks, ensuring that each agent moves towards the same goal.

[0056] In a further embodiment of the present invention, step S200 specifically includes: Step S201: when it is determined that the currently received multi-task requires the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-task is a cooperative mode; In this step embodiment, regarding the identification of common goals in the multi-agent system, specifically, first it is necessary to identify and determine whether the received tasks point to a common goal. For example, if it involves, for example, completing a project, achieving a specific performance standard, or satisfying the comprehensive needs of the user. When such commonality is found between multiple tasks, the system of the present invention will be ready to work together.

[0057] Regarding the optimization of shared resources, in addition to the common goal, the present invention will also evaluate whether there are shared resources that need to be optimized, including materials, time, space and other resources. For example, in a team project, different agents may need to share tools or information to improve overall efficiency. In this case, effective cooperation between agents will be crucial.

[0058] Regarding the determination of the cooperation mode, in the embodiment of the present invention, when the common goals and shared resources between the tasks are identified for the currently received multiple tasks, the cooperation mode will be determined as the cooperation mode, which facilitates the subsequent control of the intelligent agents to work together through communication and coordination to achieve these goals. This mode promotes information sharing, task allocation and resource integration, ensuring that each intelligent agent contributes in the field in which it excels.

[0059] In this step, through the cooperative mode, the agents can work together to solve problems, pool their respective strengths, and improve the success rate of completing tasks, especially in complex tasks or tasks that require multiple steps.

[0060] Step S202: when it is determined that the currently received multi-task requires the agents to compete for limited resources or pursue individual benefit maximization, it is determined that the cooperation mode required by the currently received multi-task is a competition mode; In this step, it is described that in a multi-agent system, when the present invention determines that the received multiple tasks involve competition for limited resources or pursuit of maximization of individual interests between agents, the collaboration mode will be determined as a competition mode.

[0061] Regarding the identification of competition for limited resources: Specifically, when the multi-agent system receives multiple tasks involving limited resources (such as funds, equipment, time, space, etc.), the present invention will automatically identify the scarcity of these resources. At this time, the behavior between the agents will become competitive, because each agent hopes to obtain more resources to improve its own goal achievement rate.

[0062] In competition mode, the controlling agent focuses on its own interests rather than the overall performance of the team. Each agent will adopt a strategy to optimize its own goals as much as possible. For example, in market competition, auction scenarios, or inventory management, the agent will analyze the behavior of competitors and make decisions to obtain higher profits or resources.

[0063] In the embodiment of the present invention, once the competitive characteristics between tasks are identified, the multi-agent system will decide to adopt a competitive mode. In this mode, the control agents will compete with each other through competition mechanisms (such as bidding, game strategies, etc.) to strive to achieve their own maximum interests, which may also affect the resource allocation of the entire system.

[0064] For example, in a complex network data transmission application environment, multiple user terminals need to compete for limited bandwidth resources. The present invention will adjust by determining the competition mechanism, and each user terminal will make a bandwidth request according to its own needs to achieve a higher data transmission speed as much as possible. In this process, the intelligent agent optimizes the use of network resources through continuous adjustment and feedback.

[0065] In this way, in a multi-agent system, when the task involves resource competition or maximizing individual interests, the determination of the competition mode not only helps to optimize resource allocation, but also motivates agents to improve efficiency and innovation. This model is applicable to a variety of scenarios, in which each agent will actively seek its own maximum interests.

[0066] Step S203: when it is determined that the currently received multi-task requires the agents to cooperate in some cases and to compete in other cases, it is determined that the cooperation mode required by the currently received multi-task is a mixed mode; In the multi-agent system of the present invention, when it is determined that some of the received multiple tasks require cooperation, while other tasks require competition, the collaboration mode will be determined to be a mixed mode.

[0067] Specifically, when the multi-agent system of the present invention receives a task, it first needs to analyze the nature of the task and determine whether it is diverse. Some tasks may require agents to work together to achieve a shared goal, such as working together in a large project; while other tasks may involve competing for limited resources or competitive interests, which requires agents to exhibit competitive behavior in certain situations.

[0068] The conditions for judging cooperation and competition in the present invention are as follows: in the hybrid mode, the system will decide when the agents need to cooperate and when they need to compete according to the changes in the specific situation. This judgment is usually based on the task objectives, resource status and environmental factors. For example, when performing a task, the agents may need to solve the problem together, but they need to compete for resources independently during the resource allocation stage.

[0069] Once the cooperative and competitive characteristics of the task are identified, the multi-agent system of the present invention will decide to adopt a hybrid mode. This mode allows the agents to flexibly adjust their behavior in different situations, achieve cooperation to achieve common goals, and also show competition when necessary to compete for personal interests or scarce resources.

[0070] Taking the application of the method of the present invention to the intelligent transportation system as an example, in urban intelligent traffic management, various traffic signals and monitoring equipment need to work together as multiple intelligent agents to optimize the overall traffic flow. During peak hours, these intelligent devices need to adjust the traffic lights together to reduce congestion. However, when the resources of a particular lane (such as parking spaces or right of way) are limited, vehicles or systems will show competitive behavior and compete for limited traffic resources. This flexible mode that combines cooperation and competition ensures the efficiency of traffic management.

[0071] As can be seen from the above, in the multi-agent system of the present invention, when the task requires the combination of cooperation and competition in different situations, the use of a hybrid mode can effectively promote the flexibility and overall efficiency of the system. This flexible strategy is applicable to a variety of application scenarios and brings greater value to the agent. In addition, the hybrid mode enables the multi-agent system to flexibly adjust strategies according to environmental changes and task requirements, thereby improving overall adaptability. Moreover, by reasonably combining cooperation and competition, the agent can optimize resource allocation, while enhancing teamwork and maximizing individual benefits.

[0072] Step S204: when it is determined that the currently received multi-task requires the agent to give priority to and achieve its own interests or goals, the self-interest mode required by the currently received multi-task is determined; In the multi-agent system of the embodiment of the present invention, when it is determined that the received multiple tasks require the agents to give priority to their own interests or goals, the collaboration mode will be determined to be the selfish mode. Specifically, in the multi-agent environment of the embodiment of the present invention, it is detected that some tasks may be self-oriented in nature, that is, it is detected that the focus of each agent is to maximize its own efficiency and benefits. This situation usually occurs in the context of limited resources and fierce competition, such as each agent hopes to obtain the highest possible benefits in a certain task.

[0073] In the self-interest mode, the agent's behavior will be based primarily on how to most effectively achieve its own goals, which may include acquiring resources, improving efficiency, increasing revenue, or achieving specific independent goals. When making decisions, the agent will prioritize analyzing which actions will help it achieve the best results, rather than focusing on the goals of the overall team or the needs of other agents.

[0074] Therefore, in the embodiment of the present invention, once the self-directed nature of the current task is identified, a decision is made to adopt a self-interest mode. In this mode, the agents will take actions to achieve their own best interests, which may lead to reduced cooperation between them, because each agent's decision is focused on its own goals rather than the overall interests of the team.

[0075] For example, in a highly competitive human-machine game, when the user character and the machine character player jointly form a team, they may show selfish behavior in order to obtain more loot or upgrade their levels. For example, after completing a dungeon task, the user character and the machine character player do not completely follow the principle of teamwork when obtaining rewards, but give priority to the benefits they can obtain. In this process, although there is sometimes cooperation, the main interest driver is still the realization of personal goals.

[0076] As can be seen from the above, in the present invention, when the tasks in the multi-agent system require the agents to give priority to their own interests, the adoption of the egoistic mode can effectively promote the motivation and efficiency of the agents. In this mode, although there may be minimal cooperation, the behavior and decision-making focus of each agent is on how to achieve its own goals. This mode is applicable to a variety of scenarios and can promote the maximization of efficiency and results. And in the egoistic mode, each agent is strongly self-motivated, driving it to continuously seek efficiency and innovation, thereby promoting the realization of their respective goals. And when each agent focuses on its own goals, they can respond quickly to environmental changes, maintain their own advantages in competition, and adapt to rapidly changing market demands.

[0077] Step S300, controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks.

[0078] According to the description of the previous steps, the multi-agent system of the embodiment of the present invention has multiple potential cooperation modes, including cooperation, competition, mixed mode and selfish mode. The present invention determines the most suitable cooperation mode according to the received task requirements, the capabilities of the agents and the environmental conditions. For example, when the task requires multiple agents to work together, the cooperation mode will be selected, and when resources are scarce, the competition mode may be switched.

[0079] In the embodiments of the present invention, once a suitable collaboration mode is determined, the agents will be guided to perform tasks according to this mode. Different agents will perform actions according to the determined new strategies and rules to achieve the task objectives. Whether it is to cooperate to achieve a common goal or to optimize their respective performance through competition, the actions of the agents will revolve around the selected mode.

[0080] In addition, in the embodiment of the present invention, when each agent detects the adjustment of task priority and the change of resource availability during the task, the control re-evaluates and switches to a collaborative mode that matches the changed tasks and resources; that is, the multi-agent system of the present invention needs to have the ability to dynamically adjust. During the task, if the situation changes (such as task priority adjustment and resource availability change), the system can also respond quickly, re-evaluate and switch to a more appropriate collaborative mode to ensure the efficient completion of the task.

[0081] Taking drone swarm monitoring as an example, when drone swarms conduct environmental monitoring, the present invention can select a suitable collaboration mode according to the nature of the task (such as monitoring range, target type). For example, when a wide area needs to be covered, a cooperative mode can be selected to mobilize all drones to form a coordinated flight to improve monitoring efficiency. When certain drones need to prioritize tracking of specific targets (such as emergencies), the present invention can be adjusted to a self-interest mode, allowing related drones to compete in these areas to obtain information and maintain attention.

[0082] Another example is an intelligent robot production line. In the production process, multiple robots can work together to perform assembly tasks. The tasks may involve different parts and processes. In collaborative mode, the robots will work together to complete the assembly. However, in the final step, in order to improve production efficiency, some robots may need to show competitiveness and focus resources on the highest priority tasks to complete the production goals on time.

[0083] Another example is intelligent traffic management. In the city's intelligent traffic system, traffic lights and traffic monitoring systems need to adjust their collaboration mode according to real-time traffic conditions. For example, during the morning rush hour, the system uses a cooperative mode to coordinate the lights to improve overall traffic flow. However, if an accident occurs on a certain road section, the system switches to a competitive mode and independently optimizes the signal light settings on that road section to reduce congestion and improve traffic efficiency.

[0084] The following is a specific application example to further explain the multi-agent application system processing method of the embodiment of the present invention: like Figure 2 As shown, the multi-agent application system implementation processing method of this specific application embodiment includes the following steps: S10, start; and enter S20; S20, the intelligent agent is initialized and enters S30; S30, select the agent mode according to the current multi-task requirements; if the cooperative mode is selected, enter S31; if the competitive mode is selected, enter S32; if the mixed mode is selected, enter S33; if the selfish mode is selected, enter S34; S31. When the corresponding collaboration mode is adopted as the cooperation mode, control the agents to collaborate and work together to achieve a common goal or optimize the use of a shared resource. The execution steps include: S311, control the effective communication and information sharing between the agents, and share each other's status, intentions and action plans; S312, control each intelligent agent to make autonomous decisions when faced with problems, and formulate appropriate cooperation strategies to allocate resources, divide tasks, and resolve conflicts; S313, control each intelligent agent to coordinate their actions according to the shared information and cooperation strategy to achieve the common goal; then enter S40.

[0085] That is, in the embodiment of the present invention, in the cooperation mode, the agents work together to achieve a common goal or optimize the use of a shared resource. The implementation process generally includes the following steps: 1) Communication and information sharing: Effective communication and information sharing are required between intelligent agents so that they can understand each other’s status, intentions, and action plans.

[0086] 2) Decision-making and cooperation strategies: Each agent needs to make autonomous decisions and develop appropriate cooperation strategies when faced with a problem, which may involve issues such as resource allocation, task division, and conflict resolution.

[0087] 3) Coordinated actions: Agents coordinate their actions based on shared information and cooperation strategies to achieve common goals; For example, taking the example of a self-driving fleet, let’s assume there is a fleet of self-driving cars (agents) whose goal is to transport passengers safely and efficiently in an urban environment.

[0088] In executing step S311: Effective Communication and Information Sharing, each autonomous vehicle (agent) needs to share their state information (such as speed, position), intention (for example, planned driving path, expected parking point) and action plan (such as scheduled lane change plan). For example, when an autonomous vehicle (agent) needs to change lanes, it sends a signal to other autonomous vehicles (agents) to inform them of its intention in advance. After receiving the information, other autonomous vehicles (agents) can adjust their speed and position to successfully complete the lane change and avoid potential collisions.

[0089] When executing step S312: Autonomous decision-making and cooperation strategy, if there is an unexpected obstacle ahead during driving (such as a bus suddenly stops), each autonomous vehicle (agent) needs to make independent judgments and respond. At this time, the autonomous vehicle (agent) will evaluate the current traffic conditions and formulate a cooperation strategy. For example, one car decides to slow down, while another car chooses to change lanes. By sharing information, the vehicles clarify their respective task division and resource allocation, such as which car is responsible for directing traffic on the obstructed section to a specific secondary road.

[0090] When executing step S313: Coordinate actions to achieve common goals, finally, after each vehicle has shared information and developed a cooperative strategy, they need to coordinate their actions to ensure that all vehicles can smoothly pass through complex traffic situations. For example, when a vehicle starts to turn right, other vehicles will adjust their driving paths according to its actions to ensure that the overall traffic is smooth and safe. This collaboration enables the entire fleet to reach the destination efficiently and achieve a common goal, which is to ensure the safety and comfort of passengers.

[0091] Through the above application examples, we can see how to enable various intelligent agents to work together to achieve goals in a cooperative mode through effective communication, autonomous decision-making and coordinated actions. Such a mechanism not only improves efficiency but also enhances security.

[0092] S32. When the corresponding collaboration mode is adopted as the competition mode, the control agents compete for limited resources or pursue the maximization of individual interests. The execution steps include: S321, controlling each intelligent agent to learn the opponent's strategy and behavior pattern, and modeling the opponent to predict its behavior and decision; S322, controlling each intelligent agent to set a corresponding competition strategy for pursuing individual interests, including selecting optimized actions, adjusting strategies, and predicting the behavior of opponents; S323, control each intelligent agent to obtain better results in resource competition by optimizing competition strategy; then enter S40.

[0093] That is, in the embodiment of the present invention, in the competition mode, each intelligent agent competes for limited resources or pursues the maximization of individual interests. The implementation process includes: A. Adversary Modeling: The agent needs to accurately model the opponent to predict its behavior and decisions, which can be achieved by learning the opponent’s strategies and behavior patterns.

[0094] B. Competitive Strategy: Agents design appropriate competitive strategies to pursue individual interests, including choosing the best actions, adjusting strategies, and predicting the behavior of opponents.

[0095] C. Resource competition: The intelligent agent achieves better results in resource competition by optimizing the competition strategy.

[0096] In the competition mode, it can be used in the processing and control of human-computer games to see how each character agent competes for limited resources by learning the opponent's behavior, setting and optimizing competition strategies. The flexible application of these strategies can help them achieve better results in the competition.

[0097] S33. When the corresponding cooperation mode is a mixed mode, the control agent needs to cooperate in some cases and show competition characteristics in other cases. The execution steps include: S331, setting each intelligent agent to have situational awareness capability to identify when cooperation and competition are needed; S332, controlling each intelligent agent to dynamically adjust cooperation or competition strategies according to environmental changes and task requirements; S333, control each intelligent agent to dynamically cooperate and compete according to the actual situation during the task execution, so as to adapt to the ever-changing environment and goals; then enter S40.

[0098] In the embodiment of the present invention, the hybrid mode combines the characteristics of cooperation and competition. The agents need to cooperate in some cases and show the characteristics of competition in other cases. The implementation process is as follows: 1) Situational awareness: Intelligent agents need to be able to recognize when to cooperate and when to compete, which requires intelligent agents to have situational awareness.

[0099] 2) Flexible strategy adjustment: The agent flexibly adjusts cooperation or competition strategies according to environmental changes and task requirements.

[0100] 3) Dynamic collaboration and competition: When executing tasks, intelligent agents dynamically collaborate or compete based on actual conditions to adapt to the ever-changing environment and goals.

[0101] Regarding the hybrid mode, let’s take the example of a drone swarm performing a search and rescue mission. When a group of drones (agents) are deployed to perform a search and rescue mission, they need to switch between cooperation and competition during the process. The specific steps are as follows: Corresponding step S331: Setting situational awareness capabilities; each drone can be equipped with environmental sensors and data analysis capabilities to identify when cooperation is needed and when independent action is required when performing a mission. In some cases, when multiple drones receive similar search missions, they need to cooperate to cover an area more efficiently, such as jointly searching for missing persons in an area. In other cases, if a drone finds a target in a critical location, such as a trapped survivor, it may need to act quickly to compete for the priority response rights of other drones.

[0102] Corresponding step S332: Dynamically adjust the cooperation or competition strategy; when the drone identifies a situation that requires cooperation, such as multiple drones discovering a similar target area, they will jointly formulate a search strategy and divide the work to cover a larger area. For example, several drones will divide the search area in a grid pattern. When the present invention detects that a drone has discovered a trapped person alone, it will control all drones (intelligent agents) to evaluate the situation and dynamically adjust the strategy so that the drone that discovers the target can take priority, and other drones may temporarily switch to a supporting role to ensure that resources are concentrated on the most urgent rescue.

[0103] Corresponding to S333: Dynamic collaboration and competition: that is, in the process of performing tasks, drones continuously adjust their behaviors according to real-time data and the current environment. For example, when a drone successfully locates and confirms the location of the trapped person, it will immediately send a signal to request support from other drones. If other drones judge that it takes time to reach the location, they can choose to give priority to providing subsequent navigation support and drone groups to search and rescue around the target area to prevent interference from other drones. In this case, accurate situational awareness and dynamic strategy adjustment enable the drone group to flexibly adapt to changing environments and goals.

[0104] Finally, the subsequent step (S40) is entered. When the drone completes the mission, whether it is a single rescue or a collective cooperation, the data will be summarized, evaluated and learned in order to improve the efficiency and effectiveness of mission execution in the future.

[0105] Through the above examples, we can clearly see how, in the hybrid mode, each intelligent agent can better adapt to the ever-changing tasks and environment through situational awareness, dynamic strategy adjustment, and real-time collaboration and competition.

[0106] S34. When the corresponding collaboration mode is the self-interest mode, the control agent gives priority to and realizes its own interests or goals. The execution steps include: S341, controlling each intelligent agent to optimize individual goals when performing tasks and interacting with other intelligent agents; S342, controlling each intelligent agent to optimize resource allocation according to resource utilization; S343, control each intelligent agent to make independent decisions according to its own needs; then enter S40.

[0107] In the embodiment of the present invention, in the self-interest mode, the design and behavior of the intelligent agent are to give priority to and achieve its own interests or goals. The implementation process includes: Individual goal optimization: Agents are primarily concerned with optimizing their own goals and interests, and even when interacting with other agents, it is based on how these interactions help them achieve their own goals.

[0108] Optimal resource allocation: When resources are limited, the agent prioritizes how to allocate resources to maximize its own interests.

[0109] Independent Decision Making: Agents make independent decisions and are not influenced by other agents unless such influence is beneficial to their individual goals.

[0110] S40, task completed, and enter S50; S50, end.

[0111] As can be seen from the above, the embodiments of the present invention can not only improve the efficiency of task completion, but also improve the system adaptability and resource utilization by controlling the multi-agent system to adopt a suitable collaborative mode to perform multiple tasks. This strategy can be effectively applied in many fields to promote the development and innovation of the intelligent system. In addition, the present invention has the following advantages: 1) Improved task execution efficiency: By selecting a suitable collaboration mode, the multi-agent system can optimize resource allocation and task execution process, and improve the overall execution efficiency.

[0112] 2) Strong adaptability: In complex environments, the collaboration mode can be flexibly adjusted according to the nature of the task and external conditions, making the system more resilient and adaptable.

[0113] 3) Maximize resource utilization: After selecting the appropriate mode, limited resources can be used more effectively, avoiding resource waste and ensuring that all agents can perform at their maximum efficiency.

[0114] 4) More conducive to promoting goal achievement: Through clear behavioral norms, each intelligent agent can work more specifically towards common or individual goals, thereby improving the success rate of task completion.

[0115] Exemplary Devices like Figure 3 As shown, an embodiment of the present invention provides a multi-agent application system implementation processing system, the system comprising: A presetting module 310 is used to pre-set a multi-agent system, and set the multi-agent system to support a cooperation mode, a competition mode, a mixed mode and a self-interest mode among the agents, so as to adapt to different application scenarios and task requirements; The cooperation mode analysis and judgment module 320 is used to analyze the currently received multi-tasks when the multi-agent system receives multiple tasks; judge the cooperation mode required by the currently received multi-tasks; when it is judged that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the cooperation mode required by the currently received multi-tasks is a cooperation mode; when it is judged that the currently received multi-tasks require the agents to compete for limited resources or pursue the maximization of individual interests, it is determined that the cooperation mode required by the currently received multi-tasks is a competition mode; when it is judged that the currently received multi-tasks require the agents to cooperate in some cases and show competitive characteristics in other cases, it is determined that the cooperation mode required by the currently received multi-tasks is a mixed mode; when it is judged that the currently received multi-tasks require the agents to give priority to and achieve their own interests or goals, it is determined that the currently received multi-tasks require a selfish mode; The task execution control module 330 is used to control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks, as described above.

[0116] Based on the above embodiments, the present invention further provides an intelligent terminal, whose principle block diagram can be shown as follows: Figure 4 As shown. The intelligent terminal includes a processor, a memory, a network interface, a display screen, and a database connected through a system bus. Among them, the processor of the intelligent terminal is used to provide computing and control capabilities. The memory of the intelligent terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the intelligent terminal is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a multi-agent application system implementation processing method is implemented. The database of the intelligent terminal is used to store a multi-agent application system implementation processing program.

[0117] Those skilled in the art will understand that Figure 4 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the smart terminal to which the scheme of the present invention is applied. The specific smart terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0118] In one embodiment, a smart terminal is provided, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, and the one or more programs include instructions for performing the following operations: Pre-setting a multi-agent system, and setting the multi-agent system to support cooperation modes among agents, such as cooperation mode, competition mode, hybrid mode and egoistic mode, so as to adapt to different application scenarios and task requirements; When the multi-agent system receives multiple tasks, it analyzes the currently received multiple tasks and determines the cooperation mode required by the currently received multiple tasks; When it is determined that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-tasks is the cooperative mode; When it is determined that the currently received multi-tasks require agents to compete for limited resources or pursue individual interests maximization, the cooperation mode required by the currently received multi-tasks is determined to be a competition mode; When it is determined that the currently received multi-task requires the agents to cooperate in some cases and to compete in other cases, it is determined that the cooperation mode required by the currently received multi-task is a mixed mode; When it is judged that the currently received multi-task requires the agent to give priority to and achieve its own interests or goals, the self-interest mode required by the currently received multi-task is determined; Control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks; the details are as described above.

[0119] The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the cooperative mode, the control agents work together to achieve a common goal or optimize the utilization of a shared resource. The execution steps include: Control effective communication and information sharing between agents, and share each other's status, intentions, and action plans; Control each agent to make autonomous decisions when faced with problems, and formulate appropriate cooperation strategies to allocate resources, divide tasks, and resolve conflicts; Control each intelligent agent to coordinate their actions based on shared information and cooperation strategies to achieve common goals.

[0120] The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the competition mode, the control agents compete for limited resources or pursue individual interests maximization. The execution steps include: Control each agent to learn the opponent's strategy and behavior pattern, and model the opponent to predict its behavior and decision; Control each agent to set a corresponding competitive strategy to pursue individual interests, including selecting optimized actions, adjusting strategies, and predicting the opponent's behavior; Control each intelligent agent to achieve better results in the competition for resources by optimizing competition strategies.

[0121] The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding cooperation mode is a mixed mode, the control agents need to cooperate in some cases and show competitive characteristics in other cases. The execution steps include: Each agent is equipped with situational awareness to identify when cooperation and competition are needed; Control each intelligent agent to dynamically adjust cooperation or competition strategies according to environmental changes and task requirements; Control each intelligent agent to dynamically collaborate and compete according to actual conditions during the execution of tasks, so as to adapt to the ever-changing environment and goals.

[0122] The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is the self-interest mode, the control agent prioritizes and realizes its own interests or goals. The execution steps include: Control each agent to optimize its individual goals when performing tasks and interacting with other agents; Control each intelligent agent to optimize resource allocation according to resource utilization; Control each intelligent agent to make independent decisions based on its own needs. The intelligent agents include: robots, drones, cameras, traffic lights and / or medical equipment.

[0123] The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multi-tasks also includes: When each agent detects an adjustment in task priority or a change in resource availability during the task, the control re-evaluates and switches to a collaborative mode that matches the changed tasks and resources, as described above.

[0124] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

Claims

1. A method for implementing a multi-agent application system, characterized in that: include: Pre-setting a multi-agent system, and setting the multi-agent system to support cooperation modes among agents, such as cooperation mode, competition mode, hybrid mode and egoistic mode, so as to adapt to different application scenarios and task requirements; When the multi-agent system receives multiple tasks, it analyzes the currently received multiple tasks and determines the cooperation mode required by the currently received multiple tasks; When it is determined that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-tasks is the cooperative mode; When it is determined that the currently received multi-tasks require agents to compete for limited resources or pursue individual interests maximization, the cooperation mode required by the currently received multi-tasks is determined to be a competition mode; When it is determined that the currently received multi-task requires the agents to cooperate in some cases and to compete in other cases, it is determined that the cooperation mode required by the currently received multi-task is a mixed mode; When it is judged that the currently received multi-task requires the agent to give priority to and achieve its own interests or goals, the self-interest mode required by the currently received multi-task is determined; Control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks.

2. The method for implementing a multi-agent application system according to claim 1, characterized in that: The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the cooperative mode, the control agents work together to achieve a common goal or optimize the utilization of a shared resource. The execution steps include: Control effective communication and information sharing between agents, and share each other's status, intentions, and action plans; Control each agent to make autonomous decisions when faced with problems, and formulate appropriate cooperation strategies to allocate resources, divide tasks, and resolve conflicts; Control each intelligent agent to coordinate their actions based on shared information and cooperation strategies to achieve common goals.

3. The method for implementing a multi-agent application system according to claim 1, characterized in that: The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is adopted as the competition mode, the control agents compete for limited resources or pursue individual interests maximization. The execution steps include: Control each agent to learn the opponent's strategy and behavior pattern, and model the opponent to predict its behavior and decision; Control each agent to set a corresponding competitive strategy to pursue individual interests, including selecting optimized actions, adjusting strategies, and predicting the opponent's behavior; Control each intelligent agent to achieve better results in the competition for resources by optimizing competition strategies.

4. The method for implementing a multi-agent application system according to claim 1, characterized in that: The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding cooperation mode is a mixed mode, the control agents need to cooperate in some cases and show competitive characteristics in other cases. The execution steps include: Each agent is equipped with situational awareness to identify when cooperation and competition are needed; Control each intelligent agent to dynamically adjust cooperation or competition strategies according to environmental changes and task requirements; Control each intelligent agent to dynamically collaborate and compete according to actual conditions during the execution of tasks, so as to adapt to the ever-changing environment and goals.

5. The method for implementing a multi-agent application system according to claim 1, characterized in that: The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks includes: When the corresponding collaboration mode is the self-interest mode, the control agent prioritizes and realizes its own interests or goals. The execution steps include: Control each agent to optimize its individual goals when performing tasks and interacting with other agents; Control each intelligent agent to optimize resource allocation according to resource utilization; Control each intelligent agent to make independent decisions based on its own needs.

6. The method for implementing a multi-agent application system according to claim 1, characterized in that: The intelligent agents include: robots, drones, cameras, traffic lights and / or medical equipment.

7. The method for implementing a multi-agent application system according to claim 1, characterized in that: The step of controlling the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks also includes: When each agent detects task priority adjustments and changes in resource availability during the task, the control re-evaluates and switches to a collaborative mode that matches the changed tasks and resources.

8. A multi-agent application system implementation processing system, characterized in that: The system comprises: A presetting module is used for A. presetting a multi-agent system, and setting the multi-agent system to support a cooperation mode, a competition mode, a mixed mode and a self-interest mode among the agents, so as to adapt to different application scenarios and task requirements; A collaboration mode analysis and judgment module is used to analyze the currently received multi-tasks when the multi-agent system receives multiple tasks; to judge the collaboration mode required by the currently received multi-tasks; when it is judged that the currently received multi-tasks require the agents to work together to achieve a common goal or optimize the use of a certain shared resource, it is determined that the collaboration mode required by the currently received multi-tasks is a collaboration mode; when it is judged that the currently received multi-tasks require the agents to compete for limited resources or pursue the maximization of individual interests, it is determined that the collaboration mode required by the currently received multi-tasks is a competition mode; when it is judged that the currently received multi-tasks require the agents to cooperate in some cases and show competitive characteristics in other cases, it is determined that the collaboration mode required by the currently received multi-tasks is a mixed mode; when it is judged that the currently received multi-tasks require the agents to give priority to and achieve their own interests or goals, it is determined that the currently received multi-tasks require a selfish mode; The task execution control module is used to control the multi-agent system to adopt a corresponding collaborative mode to execute and complete the currently received multiple tasks.

9. An intelligent terminal, characterized in that: The device comprises a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, and the one or more programs include being used to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method as described in any one of claims 1 to 7.

Images