Intelligent agent multi-algorithm fusion implementation method in simulation scene
By introducing multi-algorithm fusion and dynamic resource configuration, dynamic selection and switching decision algorithms into the agent system, the adaptability and resource waste problems of the agent system are solved, and efficient resource management and cost optimization are achieved.
Patent Information
- Application Number
- CN202510305118.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-14
- Publication Date
- 2025-08-01
AI Technical Summary
Existing agent systems can only use one or a few decision algorithms, resulting in limited adaptability and flexibility, and lack of dynamic allocation of resources for different tasks, resulting in waste of resources.
The multi-algorithm fusion method is adopted to dynamically select and switch decision algorithms through algorithm libraries, algorithm selectors, algorithm actuators and algorithm evaluators, and combine resource monitors, resource demand predictors, resource allocators and resource evaluators to dynamically allocate and adjust computing resources.
It improves the adaptability and flexibility of the agent, realizes efficient resource management, and reduces computing resources and costs.
Smart Images
Figure CN120407149A_ABST
Abstract
Description
Background Art
[0002] An agent is a computational entity with autonomy, feedback, adaptability, and sociality, which can perform various tasks in a complex environment. The core of an agent is its decision-making mechanism, that is, how to select appropriate actions according to the environmental state and goals. Different tasks and environments may require different decision-making algorithms, such as planning, learning, reasoning, gaming, etc. However, current agent systems often can only use one or a few decision-making algorithms, which limits their adaptability and flexibility. When an agent selects an algorithm, the lack of dynamic resource allocation for different tasks will cause resource waste.
[0003] Therefore, one or more methods are needed to solve the above problems.
[0004] It should be noted that the information disclosed in the above background art is only used to strengthen the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention
[0005] The purpose of the present disclosure is to provide a method, device, electronic device, and computer-readable storage medium for realizing multi-algorithm fusion of an agent in a simulation scenario, thereby at least to a certain extent overcoming one or more problems caused by the limitations and defects of related technologies.
[0006] According to one aspect of the present disclosure, there is provided a method for realizing multi-algorithm fusion of an agent in a simulation scenario, including:
[0007] A multi-algorithm fusion step, according to the adaptability of the UAV agent to tasks and the environment, the UAV agent dynamically selects and switches decision-making algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator during operation;
[0008] A dynamic resource configuration step, according to the task requirements and environmental information of the UAV agent, the UAV agent dynamically allocates and adjusts the computing resources of the processor, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator.
[0009] In an exemplary embodiment of the present disclosure, the multi-algorithm fusion step of the method further includes:
[0010] Establish a preset algorithm library, which is used to store decision-making algorithms and match a unique identifier and a parameter list for the decision-making algorithms;
[0011] Select one or more decision-making algorithms from the algorithm library according to the current state and goals of the UAV agent, and give the parameter values of the decision-making algorithms;
[0012] According to the output of the selected decision algorithm, call the decision algorithm, and use the action returned by the decision algorithm as the input of the UAV agent;
[0013] According to the action result and feedback of the UAV agent, evaluate the effect of using the decision algorithm, and feedback the evaluation result to the algorithm selector.
[0014] In an exemplary embodiment of the present disclosure, the method further includes:
[0015] The algorithm library includes decision algorithms such as the A* search algorithm, reinforcement learning algorithm, fuzzy logic algorithm, and genetic algorithm;
[0016] The identifier of the decision algorithm is used to implement the call of the decision algorithm;
[0017] The parameter list of the decision algorithm is used to specify the parameter input of the decision algorithm.
[0018] In an exemplary embodiment of the present disclosure, the dynamic resource allocation step of the method further includes:
[0019] Based on the resource monitor, monitor the task requirements of the UAV agent and the environmental information of the UAV agent;
[0020] Based on the information monitored by the resource monitor, the resource demand predictor predicts the resource demand of the UAV agent and generates resource prediction information;
[0021] Based on the resource prediction information, the resource allocator allocates computing resources from the resource pool to the UAV agent, adjusts and recovers resources, and feeds the actual result back to the resource evaluator;
[0022] According to the task result and feedback of the UAV agent, the resource evaluator evaluates the effect of the resources used and feeds the evaluation result back to the resource allocator.
[0023] In an exemplary embodiment of the present disclosure, the method further includes:
[0024] Based on the resource monitor, monitor the task requirements of the UAV agent, including task type, task complexity, task priority, task progress, and task result, as well as the environmental information of the UAV agent, including environmental state, environmental change, and environmental uncertainty.
[0025] In one aspect of the present disclosure, there is provided an intelligent agent multi-algorithm fusion implementation device in a simulation scenario, including:
[0026] A multi - algorithm fusion module, which is used to dynamically select and switch decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator according to the adaptability of the UAV agent to tasks and the environment when the UAV agent is running;
[0027] A dynamic resource configuration module, which is used to dynamically allocate and adjust computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator according to the task requirements and environmental information of the UAV agent.
[0028] In an exemplary embodiment of the present disclosure, the multi - algorithm fusion module of the device further includes:
[0029] An algorithm library, which is used to store decision algorithms and match a unique identifier and a parameter list for each decision algorithm;
[0030] An algorithm selector, which is used to select one or more decision algorithms from the algorithm library according to the current state and goals of the UAV agent and give the parameter values of the decision algorithms;
[0031] An algorithm executor, which is used to call the decision algorithm according to the output of the selected decision algorithm and use the action returned by the decision algorithm as the input of the UAV agent;
[0032] An algorithm evaluator, which is used to evaluate the effect of using the decision algorithm according to the action results and feedback of the UAV agent and feedback the evaluation results to the algorithm selector.
[0033] In an exemplary embodiment of the present disclosure, the dynamic resource configuration module of the device further includes:
[0034] A resource monitor, which is used to monitor the task requirements of the UAV agent and the environmental information where the UAV agent is located;
[0035] A resource demand predictor, which is used to predict the resource demands of the UAV agent based on the information monitored by the resource monitor and generate resource prediction information;
[0036] A resource allocator, which is used to allocate computing resources from the resource pool to the UAV agent based on the resource prediction information, adjust and recycle resources, and feedback the actual results to the resource evaluator;
[0037] A resource evaluator, which is used to evaluate the effect of the used resources according to the task results and feedback of the UAV agent and feedback the evaluation results to the resource allocator.
[0038] In one aspect of the present disclosure, there is provided an electronic device, including:
[0039] A processor; and
[0040] A memory having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by the processor, implement the method according to any one of the above items.
[0041] In one aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method according to any one of the above items is implemented.
[0042] An exemplary embodiment of the present disclosure discloses a method for implementing multi-algorithm fusion in a simulation scenario. The method includes: a multi-algorithm fusion step, in which the UAV agent dynamically selects and switches decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator at runtime, according to the UAV agent's adaptability to the task and environment; and a dynamic resource allocation step, in which the UAV agent dynamically allocates and adjusts computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator, according to the UAV agent's task requirements and environmental information. The present disclosure enables the UAV agent to dynamically select and switch between different decision algorithms during operation in a simulation environment to adapt to different tasks and environments. A dynamic resource allocation function is also incorporated, which dynamically allocates and adjusts computing resources, such as processors, memory, and bandwidth, based on the UAV agent's task requirements and environmental changes. This allows the agent to dynamically allocate and adjust computing resources at runtime to adapt to different task requirements and environmental changes. Based on the simulation results, the algorithm execution strategy for the UAV agent is derived, which also saves resources and costs.
[0043] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] The above and other features and advantages of the present disclosure will become more apparent by describing in detail example embodiments thereof with reference to the accompanying drawings.
[0045] Figure 1 A flowchart of a method for implementing multi-algorithm fusion of intelligent agents in a simulation scenario according to an exemplary embodiment of the present disclosure is shown;
[0046] Figure 2 A schematic diagram of a method for implementing multi-algorithm fusion of intelligent agents in a simulation scenario according to an exemplary embodiment of the present disclosure is shown;
[0047] Figure 3Schematic diagram of a method for multi - algorithm fusion of an agent in a simulation scenario according to an exemplary embodiment of the present disclosure;
[0048] Figure 4 Schematic diagram of dynamic resource allocation for a method for multi - algorithm fusion of an agent in a simulation scenario according to an exemplary embodiment of the present disclosure;
[0049] Figure 5 Block diagram of a device for multi - algorithm fusion of an agent in a simulation scenario according to an exemplary embodiment of the present disclosure;
[0050] Figure 6 Block diagram of an electronic device schematically shown according to an exemplary embodiment of the present disclosure;
[0051] Figure 7 Schematic diagram of a computer - readable storage medium schematically shown according to an exemplary embodiment of the present disclosure. Detailed implementation manners
[0052] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. Like reference numerals in the figures denote like or similar parts, and thus their repeated description will be omitted.
[0053] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, materials, devices, steps, etc. may be adopted. In other cases, well - known structures, methods, devices, implementations, materials, or operations are not shown or described in detail to avoid obscuring aspects of the present disclosure.
[0054] The block diagrams shown in the drawings are only functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or in one or more software - hardened modules, or in different networks and / or processor devices and / or microcontroller devices.
[0055] In the present example embodiment, first, a method for multi - algorithm fusion of an agent in a simulation scenario is provided; refer to Figure 1As shown in , the method for realizing multi - algorithm fusion of agents in a simulation scenario may include the following steps:
[0056] Multi - algorithm fusion step S110: According to the adaptability of the UAV agent to tasks and the environment, the UAV agent dynamically selects and switches decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator during operation.
[0057] Dynamic resource allocation step S120: According to the task requirements and environmental information of the UAV agent, the UAV agent dynamically allocates and adjusts computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator.
[0058] A method for realizing multi - algorithm fusion of agents in a simulation scenario in an exemplary embodiment of the present disclosure, wherein the method includes: a multi - algorithm fusion step. According to the adaptability of the UAV agent to tasks and the environment, the UAV agent dynamically selects and switches decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator during operation; a dynamic resource allocation step. According to the task requirements and environmental information of the UAV agent, the UAV agent dynamically allocates and adjusts computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator. The present disclosure enables the UAV agent to dynamically select and switch different decision algorithms during the operation in a simulation environment to adapt to different tasks and environments. At the same time, a dynamic resource allocation function is added. This function can dynamically allocate and adjust computing resources according to the task requirements and environmental changes of the UAV agent, such as processors, memory, bandwidth, etc. In this way, the agent can dynamically allocate and adjust computing resources during operation to adapt to different task requirements and environmental changes. Based on the simulation results, the algorithm operation strategy of the UAV agent is obtained, and resources and costs can also be saved.
[0059] Next, a method for realizing multi - algorithm fusion of agents in a simulation scenario in the present exemplary embodiment will be further described.
[0060] Embodiment 1:
[0061] In the embodiment of this example, for the method of multi - algorithm fusion of the UAV agent of the present disclosure, in the established simulation scenario of UAV use, after dynamically monitoring the operation state of the UAV agent through running this method, resource adjustment and deployment are carried out to achieve the adaptability and flexibility of the UAV agent.
[0062] Specifically, the following effects can be achieved:
[0063] 1. Improve the adaptability and autonomy of the UAV agent: The multi-algorithm fusion method can dynamically select and switch the most suitable decision algorithm from the algorithm library according to the state and goals of the UAV agent.
[0064] 2. Achieve the efficiency and robustness of the UAV agent: The system provides a way of dynamic resource allocation, which can dynamically allocate and adjust computing resources according to the task requirements and environmental changes of the UAV agent.
[0065] 3. Improve the cost performance and competitiveness of the system: Through the effective coordination of resources, it realizes the saving of computing resources and events, and reduces the cost and overhead.
[0066] In the multi-algorithm fusion step S110, according to the adaptability of the UAV agent to the task and environment, the UAV agent can dynamically select and switch decision algorithms based on the algorithm library, algorithm selector, algorithm executor, and algorithm evaluator during operation.
[0067] In the embodiment of this example, the multi-algorithm fusion step of the method further includes:
[0068] Establish a preset algorithm library, which is used to store decision algorithms and match a unique identifier and a parameter list for each decision algorithm;
[0069] Select one or more decision algorithms from the algorithm library according to the current state and goals of the UAV agent, and give the parameter values of the decision algorithms;
[0070] According to the output of the selected decision algorithm, call the decision algorithm and use the action returned by the decision algorithm as the input of the UAV agent;
[0071] According to the action result and feedback of the UAV agent, evaluate the effect of using the decision algorithm and feedback the evaluation result to the algorithm selector.
[0072] In the embodiment of this example, the method further includes:
[0073] The algorithm library includes decision algorithms of A* search algorithm, reinforcement learning algorithm, fuzzy logic algorithm, and genetic algorithm;
[0074] The identifier of the decision algorithm is used to implement the call of the decision algorithm;
[0075] The parameter list of the decision algorithm is used to specify the parameter input of the decision algorithm.
[0076] In the dynamic resource allocation step S120, according to the task requirements and environmental information of the UAV agent, the UAV agent dynamically allocates and adjusts the computing resources of the processor, memory, and bandwidth based on the resource monitor, resource demand predictor, resource allocator, and resource evaluator.
[0077] In the embodiment of this example, the dynamic resource allocation step of the method further includes:
[0078] Based on the resource monitor, monitor the task requirements of the UAV agent and the environmental information where the UAV agent is located;
[0079] Based on the information monitored by the resource monitor, the resource demand predictor predicts the resource demand of the UAV agent and generates resource prediction information;
[0080] Based on the resource prediction information, the resource allocator allocates computing resources from the resource pool to the UAV agent, adjusts and recycles resources, and feeds the actual results back to the resource evaluator;
[0081] According to the task results and feedback of the UAV agent, the resource evaluator evaluates the effect of the resources used and feeds the evaluation results back to the resource allocator.
[0082] In the embodiment of this example, the method further includes:
[0083] Based on the resource monitor, monitor the task requirements of the task type, task complexity, task priority, task progress, and task results of the UAV agent, as well as the environmental information of the environmental state, environmental changes, and environmental uncertainties where the UAV agent is located.
[0084] In the embodiment of this example, the purpose of the present invention is to provide a flexible and highly adaptable multi-algorithm fusion method for UAV agents, and dynamically deploy and adjust resources to meet the requirements of the high efficiency and robustness of UAV agent computing, and effectively manage the adaptability of UAV agents to different tasks and environments.
[0085] Embodiment 2
[0086] In the embodiment of this example, as Figure 2 shown, the principle and method of the dynamic resource allocation of the present disclosure, this method calls the optimization of the hardware resource support software algorithm layer. This method can dynamically allocate and adjust computing resources, such as processors, memory, bandwidth, etc., according to the task requirements and environmental changes of the UAV agent. This method includes a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator. The present invention can improve the adaptability and flexibility of UAV agents, and at the same time can save computing resources and costs.
[0087] In the embodiments of this example, the implementation of the multi-algorithm fusion method includes:
[0088] First, a three-dimensional model of the drone in the actual use scenario is established, and a simulation scenario of the drone is generated based on this three-dimensional model. In this scenario, it is assumed that the drone agent needs to fly to a specified location. The drone agent can use the multi-algorithm fusion method provided by the present invention, as Figure 1 shown. This method includes four steps:
[0089] 1. Select a suitable algorithm.
[0090] The agent inputs the current goal and state to be achieved into the algorithm selector. The algorithm selector selects one or more suitable decision algorithms from the algorithm library based on the current state and gives the corresponding parameter values. There are four available decision algorithms stored in the algorithm library, including the A* search algorithm, reinforcement learning algorithm, fuzzy logic algorithm, and genetic algorithm. Each algorithm has a unique identifier and a parameter list. The unique identifier is used to accurately call the relevant algorithm later, and the parameter list is used to specify the parameter input of the algorithm. For example, A* search needs to specify the heuristic function and cost function, and reinforcement learning needs to specify the reward function and learning rate, etc. When the drone agent first enters flight, it may select A* search as the initial decision algorithm and specify the heuristic function as the Euclidean distance and the cost function as the number of moving steps; when the drone agent encounters an obstacle or a dead end, it may switch to reinforcement learning as an alternative decision algorithm and specify the reward function as the change in distance to the destination and the learning rate as 0.1, etc.;
[0091] 2. Execute the algorithm
[0092] According to the output of the algorithm selector, the corresponding decision algorithm is called, and the action returned by it is used as the output of the drone agent. For example, when the drone agent uses A* search, it will calculate the shortest path from the current position to the exit according to the heuristic function and cost function and select the next action along this path; when the drone agent uses reinforcement learning, it will update its action value function according to the reward function and learning rate and select the action with the highest value;
[0093] 3. Evaluate the algorithm
[0094] After the UAV agent executes the corresponding task, it evaluates the effectiveness of the decision-making algorithm used based on the action results and feedback of the UAV agent, and feeds back the evaluation results to the algorithm selector. For example, when the UAV agent uses A* search, if it finds that the path is blocked by an obstacle or is a dead end, it will evaluate the effectiveness of this decision-making algorithm as low and notify the algorithm selector to switch to other decision-making algorithms; when the agent uses reinforcement learning, if it finds that this action makes it closer to the destination, it will evaluate the effectiveness of this decision-making algorithm as high and notify the algorithm selector.
[0095] 4. Feedback Results
[0096] After the agent calls the decision-making algorithm, it will input the corresponding effectiveness evaluation into the algorithm selector to facilitate better algorithm switching. The algorithm selector can better recommend and select appropriate algorithms based on past experience, and input them to the algorithm executor for execution, enabling the UAV agent to achieve better execution effects.
[0097] In this way, the UAV agent can dynamically select and switch different decision-making algorithms during operation to adapt to different tasks and environments. This can improve the adaptability and flexibility of the UAV agent, and at the same time save computing resources and time.
[0098] In the embodiment of this example, the implementation of the dynamic resource allocation method includes:
[0099] On this basis, the dynamic resource allocation method is added. The dynamic resource allocation method refers to dynamically adjusting according to the resources required by the task when the UAV agent executes the algorithm. As Figure 3 shown, this method belongs to the hardware resource layer, and can support the specific implementation of the software algorithm layer through dynamic resource mobilization.
[0100] The dynamic resource allocation method includes the following four steps:
[0101] 1. The resource monitor monitors the agent
[0102] The resource monitor monitors information such as the current task type, task complexity, task priority, task progress, task results, etc. of the UAV agent, as well as information such as the environmental state, environmental changes, and environmental uncertainties of the UAV agent, so as to continuously grasp the current task execution status of the UAV agent.
[0103] 2. The resource demand predictor predicts resource demands
[0104] The resource demand predictor predicts the resource demands of the UAV agent in a future period of time based on the information monitored by the resource monitor, such as the number of required processors, memory size, bandwidth speed, etc., so as to pre-deploy and plan resources.
[0105] 3. The resource allocator allocates resources
[0106] According to the prediction information, the resource allocator allocates appropriate computing resources from the available resource pool to the UAV agent, and adjusts and reclaims resources according to the actual situation. And the actual results are fed back to the resource evaluator.
[0107] 4. The resource evaluator evaluates the effect
[0108] According to the task results and feedback of the UAV agent, the resource evaluator evaluates the effect of the resources used and feeds back the evaluation results to the resource allocator. The resource allocator can better allocate resources to the UAV agent to execute the algorithm based on past experience, so that the UAV agent achieves better resource utilization effects.
[0109] In this way, the UAV agent can dynamically allocate and adjust computing resources during operation to adapt to different task requirements and environmental changes. This can improve the performance and efficiency of the UAV agent, and at the same time save resources and costs.
[0110] It should be noted that although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be executed in that specific order, or that all the steps shown must be executed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.
[0111] In addition, in the present exemplary embodiment, there is also provided an apparatus for implementing multi - algorithm fusion of an agent in a simulation scenario. Referring to Figure 5 As shown, the apparatus 200 for implementing multi - algorithm fusion of an agent in a simulation scenario may include: a multi - algorithm fusion module 210 and a dynamic resource configuration module 220. Among them:
[0112] The multi - algorithm fusion module 210 is used to dynamically select and switch decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator during the operation of the UAV agent according to the adaptability of the UAV agent to tasks and the environment;
[0113] The dynamic resource configuration module 220 is used to dynamically allocate and adjust computing resources of a processor, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator according to the task requirements and environmental information of the UAV agent.
[0114] In the embodiment of the present example, as Figure 3 shown, the multi - algorithm fusion module 210 includes the following parts:
[0115] Algorithm library: Stores a variety of available decision algorithms, each algorithm having a unique identifier and a parameter list;
[0116] Algorithm selector: Based on the current state and goals of the UAV agent, selects one or more appropriate decision algorithms from the algorithm library and provides the corresponding parameter values;
[0117] Algorithm executor: According to the output of the algorithm selector, invokes the corresponding decision algorithm and uses its returned actions as the input to the UAV agent;
[0118] Algorithm evaluator: Based on the action results and feedback of the UAV agent, evaluates the effectiveness of the decision algorithm used and feeds back the evaluation results to the algorithm selector.
[0119] In the embodiment of this example, the dynamic resource configuration module 220 can dynamically allocate and adjust computing resources, such as processors, memory, bandwidth, etc., according to the task requirements and environmental changes of the agent. The dynamic resource configuration module 220 includes the following parts:
[0120] Resource monitor: Used to monitor information such as the current task type, task complexity, task priority, task progress, task results, etc. of the UAV agent, as well as information such as the environmental state, environmental changes, environmental uncertainty, etc. of the UAV agent;
[0121] Resource demand predictor: Used to predict the resource demands of the UAV agent in a future period of time based on the output of the resource monitor, such as the required number of processors, memory size, bandwidth speed, etc.;
[0122] Resource allocator: Used to allocate appropriate computing resources from the available resource pool to the UAV agent according to the output of the resource demand predictor and make adjustments and recoveries according to the actual situation;
[0123] Resource evaluator: Used to evaluate the effectiveness of the resources used based on the task results and feedback of the UAV agent and feed back the evaluation results to the resource allocator.
[0124] In this way, the agent can dynamically allocate and adjust computing resources during runtime to adapt to different task requirements and environmental changes. This can improve the performance and efficiency of the agent, and at the same time save resources and costs.
[0125] After adding the dynamic resource configuration method to the multi-algorithm fusion method of the agent, its schematic diagram is as Figure 4 shown. Dynamic resource configuration is mainly responsible for resource allocation at the hardware level and adjusts according to the running requirements of the agent.
[0126] It should be noted that although several modules or units of the intelligent agent multi-algorithm fusion implementation device 200 in the simulation scenario are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of the two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.
[0127] In addition, in an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.
[0128] Those skilled in the art can understand that various aspects of the present invention can be implemented as a system, a method, or a program product. Therefore, various aspects of the present invention can be specifically implemented in the following forms, namely: a complete hardware embodiment, a complete software embodiment (including firmware, microcode, etc.), or an embodiment combining hardware and software aspects, which can be collectively referred to as "circuit", "module", or "system" here.
[0129] Next, refer to Figure 6 to describe the electronic device 300 according to this embodiment of the present invention. Figure 6 The shown electronic device 300 is only an example and should not impose any limitation on the functions and usage scope of the embodiments of the present invention.
[0130] As Figure 6 shown, the electronic device 300 is presented in the form of a general computing device. The components of the electronic device 300 may include, but are not limited to: at least one of the above-mentioned processing units 310, at least one of the above-mentioned storage units 320, a bus 330 connecting different system components (including the storage unit 320 and the processing unit 310), and a display unit 340.
[0131] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 310, so that the processing unit 310 executes the steps according to various exemplary embodiments of the present invention described in the above "Exemplary Method" section of this specification. For example, the processing unit 310 can execute steps S110 to S120 as Figure 1 shown.
[0132] The storage unit 320 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 3201 and / or a cache storage unit 3202, and may further include a read-only storage unit (ROM) 3203.
[0133] The storage unit 320 may also include a program / utility 3204 having a set (at least one) of program modules 3205. Such program modules 3205 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.
[0134] The bus 330 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus structures.
[0135] The electronic device 300 may also communicate with one or more external devices 370 (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 300, and / or may communicate with any device that enables the electronic device 300 to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through the input / output (I / O) interface 350. Also, the electronic device 300 may communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 360. As shown in the figure, the network adapter 360 communicates with other modules of the electronic device 300 through the bus 330. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 300, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.
[0136] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or can be implemented by the way of software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
[0137] In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium, on which a program product capable of implementing the above methods in this specification is stored. In some possible embodiments, various aspects of the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary embodiments of the present invention described in the "Exemplary Method" section above in this specification.
[0138] Referring Figure 7 As shown, a program product 400 for implementing the above method according to an embodiment of the present invention is described. It may be a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.
[0139] The program product may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, but not be limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.
[0140] The computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries the readable program code. Such a propagated data signal may take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable signal medium may also be any readable medium other than the readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0141] The program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination of the above.
[0142] The program code for performing the operations of the present invention can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., via an Internet service provider through the Internet).
[0143] In addition, the above-mentioned drawings are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present invention, rather than for limiting purposes. It is easy to understand that the processes shown in the above-mentioned drawings do not indicate or limit the chronological order of these processes. Additionally, it is also easy to understand that these processes can be executed synchronously or asynchronously, for example, in multiple modules.
[0144] Those skilled in the art will readily think of other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include well-known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the claims.
[0145] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.
Claims
1. A method for realizing the fusion of multiple algorithms of an intelligent agent in a simulation scenario, characterized in that The method includes: A multi-algorithm fusion step. According to the adaptability of the UAV agent to tasks and the environment, the UAV agent dynamically selects and switches decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator during operation. A dynamic resource allocation step. According to the task requirements and environmental information of the UAV agent, the UAV agent dynamically allocates and adjusts computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator.
2. The method according to claim 1, characterized in that The multi-algorithm fusion step of the method further includes: Establishing a preset algorithm library, which is used to store decision algorithms and match a unique identifier and a parameter list for each decision algorithm. Selecting one or more decision algorithms from the algorithm library according to the current state and goals of the UAV agent, and giving the parameter values of the decision algorithms. According to the output of the selected decision algorithm, calling the decision algorithm and using the action returned by the decision algorithm as the input of the UAV agent. Evaluating the effect of using the decision algorithm according to the action result and feedback of the UAV agent, and feeding back the evaluation result to the algorithm selector.
3. The method according to claim 1, wherein The method further includes: The algorithm library includes decision algorithms such as the A* search algorithm, reinforcement learning algorithm, fuzzy logic algorithm, and genetic algorithm. The identifier of the decision algorithm is used to implement the call of the decision algorithm. The parameter list of the decision algorithm is used to specify the parameter input of the decision algorithm.
4. The method according to claim 1, characterized in that, The dynamic resource allocation step of the method further includes: Monitoring the task requirements of the UAV agent and the environmental information of the UAV agent based on a resource monitor. Based on the information monitored by the resource monitor, a resource demand predictor predicts the resource demand of the UAV agent and generates resource prediction information. Based on the resource prediction information, a resource allocator allocates computing resources from the resource pool to the UAV agent, adjusts and reclaims resources, and feeds back the actual result to the resource evaluator. According to the task result and feedback of the UAV agent, a resource evaluator evaluates the effect of the resources used and feeds back the evaluation result to the resource allocator.
5. The method according to claim 4, characterized in that, The method further includes: Monitoring the task requirements of the UAV agent, such as the task type, task complexity, task priority, task progress, and task result, and the environmental information of the UAV agent, such as the environmental state, environmental changes, and environmental uncertainty, based on a resource monitor.
6. An intelligent agent multi-algorithm fusion implementation device in a simulation scenario, characterized in that, The device includes: A multi-algorithm fusion module, which is used to dynamically select and switch decision algorithms based on an algorithm library, an algorithm selector, an algorithm executor, and an algorithm evaluator according to the adaptability of the UAV agent to tasks and the environment during operation. A dynamic resource allocation module, which is used to dynamically allocate and adjust computing resources such as processors, memory, and bandwidth based on a resource monitor, a resource demand predictor, a resource allocator, and a resource evaluator according to the task requirements and environmental information of the UAV agent.
7. The device according to claim 1, wherein The multi-algorithm fusion module of the device further includes: An algorithm library, which is used to store decision algorithms and match a unique identifier and a parameter list for each decision algorithm. An algorithm selector for selecting one or more decision algorithms from an algorithm library according to the current state and target of the UAV agent, and giving the parameter values of the decision algorithms; An algorithm executor for calling the decision algorithm according to the output of the selected decision algorithm, and taking the action returned by the decision algorithm as the input of the UAV agent; An algorithm evaluator for evaluating the effect of using the decision algorithm according to the action result and feedback of the UAV agent, and feeding back the evaluation result to the algorithm selector.
8. The device according to claim 1, characterized in that, The dynamic resource configuration module of the device further includes: A resource monitor for monitoring the task requirements of the UAV agent and the environmental information where the UAV agent is located; A resource demand predictor for predicting the resource demand of the UAV agent based on the information monitored by the resource monitor and generating resource prediction information; A resource allocator for allocating computing resources from the resource pool to the UAV agent based on the resource prediction information, adjusting and recycling resources, and feeding back the actual result to the resource evaluator; A resource evaluator for evaluating the effect of the used resources according to the task result and feedback of the UAV agent, and feeding back the evaluation result to the resource allocator.
9. An electronic device, characterized in that, Comprising A processor; and A memory storing computer-readable instructions, which when executed by the processor implement the method according to any one of claims 1 to 5.
10. A computer-readable storage medium, characterized in that, A computer program is stored thereon, which when executed by a processor implements the method according to any one of claims 1 to 5.