Service execution method and system and computing equipment

Abstract

The invention relates to the technical field of computers, provides a service execution method and system and computing equipment, and is used for effectively meeting service requirements of various complex service scenes, reducing complexity of a processing process and improving predictability and flexibility of the processing process. According to the service execution system, in response to a received service request, one or more multi-agent systems are called to process the service request, each multi-agent system comprises one or more agents, an agent assembly line is arranged in at least one multi-agent system, and the agent assembly line comprises a plurality of agents executed in sequence; and in the process of processing the service request, calling the agent assembly line.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a business execution method, system and computing device. Background Technology

[0002] A multi-agent system (MAS) is an artificial intelligence system that supports business implementation through the collaboration of multiple intelligent agents. However, when faced with highly complex tasks, the collaboration between multiple intelligent agents often involves a large amount of communication, coordination, and synchronization, resulting in a complex task processing flow and poor predictability. Summary of the Invention

[0003] This application provides a business execution method, system, and computing device to effectively meet the business needs of various complex business scenarios, reduce the complexity of the processing process, and improve the predictability and flexibility of the processing process.

[0004] To achieve the above objectives, the embodiments of this application adopt the following technical solutions:

[0005] In a first aspect, embodiments of this application provide a business execution method, comprising: responding to a received business request by invoking one or more multi-agent systems to process the business request, each multi-agent system including one or more intelligent agents, at least one multi-agent system being provided with an intelligent agent pipeline, the intelligent agent pipeline including multiple intelligent agents executed sequentially; and invoking the intelligent agent pipeline during the processing of the business request.

[0006] Based on this, an agent pipeline can be set up in a multi-agent system to achieve ordered invocation of multiple agents, reducing the complexity of the processing flow and supporting accurate prediction of the execution flow based on the agent pipeline, thus improving predictability. Furthermore, these multiple agents can be agents from one or more multi-agent systems, enabling collaboration between agents in different multi-agent systems to process business requests, demonstrating high flexibility in collaboration. Therefore, the embodiments of this application can effectively meet the business needs of various complex business scenarios, reduce the complexity of the processing process, and improve the predictability and flexibility of the processing process.

[0007] In one possible implementation, in response to a received business request, one or more multi-agent systems are invoked to process the business request, including: in response to the business request, determining the business information corresponding to the business request; and based on the business information, invoking one or more multi-agent systems to process the business request.

[0008] In one possible implementation, the business information includes a business type; based on the business information, one or more multi-agent systems are invoked to process the business request, including: determining the agent pipeline corresponding to the business type; and invoking agents in one or more multi-agent systems that are on the agent pipeline to process the business request in the order indicated by the agent pipeline.

[0009] In one possible implementation, determining the agent pipeline corresponding to the business type includes: obtaining the agent pipeline corresponding to the business type from a subscription platform; the subscription platform is used to store multiple agent pipelines.

[0010] In one possible implementation, one or more intelligent agents include a user intelligent agent; in response to a business request, determining the business information corresponding to the business request includes: in response to the business request, invoking the user intelligent agent to parse the business request and obtain the business information.

[0011] In one possible implementation, the business execution method further includes: during the processing of a business request, invoking at least one execution tool on the subscription platform through an agent.

[0012] In one possible implementation, multiple execution tools are registered on the subscription platform; one or more execution tools are configured with access rules, which are used to indicate which agents have the necessary access rights.

[0013] In one possible implementation, the business execution method further includes: sending capability information to the subscription platform; the capability information is used to indicate the intelligent agent pipeline and / or execution tools that can be invoked.

[0014] Secondly, a business execution device is provided, including: a calling unit;

[0015] The invocation unit is used to respond to a received business request by invoking one or more multi-agent systems to process the business request. Each multi-agent system includes one or more agents, and at least one multi-agent system is equipped with an agent pipeline, which includes multiple agents executed in sequence. The invocation unit is also used to invoke the agent pipeline during the processing of the business request.

[0016] In one possible implementation, the calling unit is specifically used to: respond to a business request, determine the business information corresponding to the business request; and, based on the business information, call one or more multi-agent systems to process the business request.

[0017] In one possible implementation, the business information includes a business type; and a calling unit, specifically used to: determine the agent pipeline corresponding to the business type; and, in accordance with the order indicated by the agent pipeline, call one or more agents in the agent pipeline of the multi-agent system to process the business request.

[0018] In one possible implementation, the calling unit is specifically used to: obtain the agent pipeline corresponding to the business type from the subscription platform; the subscription platform is used to store multiple agent pipelines.

[0019] In one possible implementation, one or more intelligent agents include a user intelligent agent; the invocation unit is specifically used to: in response to a business request, invoke the user intelligent agent to parse the business request and obtain business information.

[0020] In one possible implementation, the calling unit is also used to invoke at least one execution tool on the subscription platform through the agent during the processing of business requests.

[0021] In one possible implementation, multiple execution tools are registered on the subscription platform; one or more execution tools are configured with access rules, which are used to indicate which agents have the necessary access rights.

[0022] In one possible implementation, the service execution apparatus further includes a sending unit. The sending unit is used to send capability information to the subscription platform; the capability information is used to indicate available intelligent agent pipelines and / or execution tools.

[0023] Thirdly, a business execution system is provided, which is used to run on one or more computing devices, including: one or more multi-agent systems, which are used to process business requests; each multi-agent system includes one or more agents, and at least one multi-agent system is provided with an agent pipeline, which includes multiple agents executed in sequence, and the multi-agent system is used to invoke the agent pipeline in the process of processing business requests.

[0024] Fourthly, a computing device is provided. The computing device may include a memory and a processor. The memory and processor are electrically connected. The memory stores program instructions, and the processor executes the program instructions to cause the computing device to perform the business execution method of the first aspect and its possible implementations. Alternatively, the computing device may be deployed with a business execution system as described in the third aspect.

[0025] Fifthly, a chip is provided, comprising: a processor and an interface circuit; the interface circuit for receiving program instructions and transmitting them to the processor; and the processor for executing the program instructions to perform the business execution method of the first aspect and its possible implementations, or to deploy a business execution system such as the third aspect.

[0026] Sixthly, a computer-readable storage medium is provided, which stores program instructions. When the program instructions in the computer-readable storage medium are executed by a computing device, the computing device performs the business execution method of the first aspect and its possible implementations.

[0027] In a seventh aspect, a computer program product containing program instructions is provided. When the program instructions in a computer-readable storage medium are executed by a computing device, the computing device performs the business execution method of the first aspect and its possible implementations.

[0028] It should be understood that the technical effects of any of the implementation methods in aspects two through seven are the same as those of the corresponding implementation methods in aspect one, and will not be repeated here. Based on the implementation methods provided in the above aspects, the embodiments of this application can be further combined to provide more implementation methods. Attached Figure Description

[0029] Figure 1 This application provides a schematic diagram of the structure of a business execution system according to an embodiment of the present application.

[0030] Figure 2 This is a schematic diagram of the structure of another business execution system provided in an embodiment of this application;

[0031] Figure 3 This is a schematic diagram of the structure of another business execution system provided in an embodiment of this application;

[0032] Figure 4 This is a schematic diagram of the hardware structure of a computing device provided in an embodiment of this application;

[0033] Figure 5 A flowchart illustrating a business execution method provided in an embodiment of this application;

[0034] Figure 6 A flowchart illustrating a business execution process provided in an embodiment of this application;

[0035] Figure 7 This is a functional structure diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0036] In the description of the embodiments in this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Furthermore, "at least one" refers to one or more, and "multiple" refers to two or more. The terms "first," "second," etc., do not limit the quantity or execution order, and the terms "first," "second," etc., do not necessarily imply that they are different.

[0037] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0038] This application provides a business execution method that enables the orderly invocation of multiple agents by setting up an agent pipeline in a multi-agent system, reducing the complexity of the processing flow and supporting accurate prediction of the execution flow based on the agent pipeline, thus improving predictability. Furthermore, these multiple agents can be agents from one or more multi-agent systems, allowing collaboration between agents in different multi-agent systems to process business requests, demonstrating high flexibility in collaboration. Therefore, this application can effectively meet the business needs of various complex business scenarios, reducing the complexity of the processing process and improving its predictability and flexibility. The business execution method provided in this application can support the efficient utilization of large-scale artificial intelligence models through multiple multi-agent systems, improving the adaptability of large-scale artificial intelligence models to application scenarios and making it applicable to any scenario where large-scale artificial intelligence models support business implementation. These large-scale artificial intelligence models can be large language models, large visual models, or multimodal models, etc. Examples include smart factory scenarios that use large-scale AI models to coordinate and schedule production equipment, handling robots, and sensors; logistics and distribution scenarios that use large-scale AI models to dynamically allocate vehicles and packages; and enterprise scenarios that use large-scale AI models to manage business modules such as human resources and after-sales service. It should be noted that the above examples are merely illustrative and do not limit the applicability of these business execution methods.

[0039] The business execution method provided in this application can be applied to computing devices. Computing devices may include servers, etc. A server can be a single physical server, or two or more physical servers sharing different responsibilities and cooperating to achieve various server functions. For example, a server can be a blade server, a high-density server, a rack server, or a tower server, etc. Terminal devices may include personal digital assistants (PDAs), ultra-mobile personal computers (UMPCs), laptops, netbooks, desktop computers, all-in-one computers, etc. It should be noted that this application does not limit the form of the computing device; the above is merely illustrative.

[0040] The computing device may be deployed with a business execution system. This business execution system may include one or more multi-agent systems. The multi-agent systems are used to process business requests. Each multi-agent system includes one or more agents. At least one multi-agent system has an agent pipeline. The agent pipeline includes multiple agents executed sequentially. The multi-agent system is used to invoke the agent pipeline during the processing of business requests.

[0041] A multi-agent system refers to an artificial intelligence system that supports business implementation through the cooperation of multiple intelligent agents. Different multi-agent systems can be used to handle different business blocks corresponding to the business execution system.

[0042] As an example, multiple multi-agent systems in a business execution system can be deployed in a distributed architecture. Any multi-agent system can invoke the capabilities of its associated multi-agent systems, supporting flexible collaboration for efficient task execution. Associated multi-agent systems refer to multi-agent systems whose corresponding business blocks have dependencies on the business blocks corresponding to any of those multi-agent systems.

[0043] In one possible implementation, the business execution system can also be deployed and run on multiple computing devices. For example, the multiple multi-agent systems in the business execution system can be independently configured on different computing devices. Alternatively, some of the multi-agent systems in the multiple multi-agent systems can be integrated into one computing device, while others can be integrated into another. Again, some of the multi-agent systems in the multiple multi-agent systems can be integrated into one computing device, while others can be independently configured on different computing devices. This application does not impose any limitations on these embodiments.

[0044] As an example, such as Figure 1 The diagram shown is a structural schematic of a business execution system provided in an embodiment of this application. Figure 1The business execution system shown includes a first multi-agent system 100 and a second multi-agent system 200.

[0045] Figure 1 The first multi-agent system 100 is allowed to call the second multi-agent system 200. The first multi-agent system 100 is any one of multiple multi-agent systems. The second multi-agent system 200 is used to represent a multi-agent system whose corresponding business block has a dependency relationship with the business block corresponding to the first multi-agent system 100.

[0046] For example, if the business execution system corresponds to a smart factory production business, the first multi-agent system 100 can be used to handle the business block of managing production equipment, and the second multi-agent system 200 can be used to handle the business block of managing handling robots. There is a dependency between the business blocks of the two systems.

[0047] For example, if the business execution system corresponds to enterprise operation and management business, the first multi-agent system 100 can be used to handle the business block of sales management, and the second multi-agent system 200 can be used to handle the business block of market management. There is a dependency relationship between the business blocks of the two systems.

[0048] Figure 1 The first multi-agent system 100 includes a first planning agent 101 and multiple first execution agents 102. The second multi-agent system 200 includes a second planning agent 201 and multiple second execution agents 202.

[0049] It should be noted that the number and connection relationship of the multi-agent system included in the above-mentioned business execution system, as well as the planning agent and execution agent included in any multi-agent system, are only examples, and the embodiments of this application do not limit this.

[0050] Intelligent agents (such as planning agents or execution agents) are entities with independent computing and decision-making capabilities, capable of operating autonomously to a certain extent and interacting with other intelligent agents, and can be realized based on artificial intelligence (AI) models.

[0051] The planning agent (such as the first planning agent 101 or the second planning agent 201) is used to parse the task request, and to perform reasonable task orchestration based on the capabilities of the execution agents that can be scheduled, and further issue tasks to the execution agents.

[0052] An execution agent (such as a first execution agent 102 or a second execution agent 202) is used to execute tasks, and can execute assigned task steps by calling preset execution tools.

[0053] Furthermore, the first planning agent 101 can be used to respond to a business request, obtain first capability information and second capability information, and based on the first capability information and second capability information, schedule the first execution agent 101 and / or the second execution agent 201 to execute the task corresponding to the business request.

[0054] The first capability information is used to indicate the capabilities of the first executive agent 101, that is, the functions supported and implemented by the first executive agent 101 included in the first multi-agent system 100. The second capability information is used to indicate the capabilities of the second executive agent 201 in the second multi-agent system, that is, the functions supported and implemented by the second executive agent 201 included in the second multi-agent system 200.

[0055] A business request is used to represent a processing request for a specified object within the business logic of a business execution system. For example, a business request could be a production request for a specified product in a smart factory. Or, a business request could be a delivery request for a specified package in logistics and distribution. This application's embodiments do not impose such limitations.

[0056] Specifically, the first planning agent 101 can read the capabilities of each first execution agent 101 within the first multi-agent system 100 from the database corresponding to the first multi-agent system 100, and obtain the capabilities of each second execution agent 201 within the second multi-agent system 200. Furthermore, the first planning agent 101 can determine the capabilities required to process the business request based on the processing flow or type corresponding to the business request, and further select a target execution agent from the first execution agents 101 and / or the second execution agents 201 that meets the capability requirements, and schedule the target execution agent to execute the task corresponding to the business request.

[0057] In one possible implementation, the first planning agent 101 can send a capability information acquisition request to the second planning agent 201 in the second multi-agent system 200. In response to this capability information acquisition request, the second planning agent 201 can read the capabilities of each second execution agent 201 within the second multi-agent system 200 from the database corresponding to the second multi-agent system 200, and send the read information, along with second capability information, to the first planning agent 101. Accordingly, the first planning agent 101 can receive the second capability information from the second planning agent 201.

[0058] In one possible implementation, combining Figure 1 The business execution system shown is as follows: Figure 2 The diagram shown is a structural schematic of another business execution system provided in an embodiment of this application. The business execution system may include a multi-agent system group, storage, resources, tools, and infrastructure.

[0059] A multi-agent system group comprises multiple multi-agent systems. When a business execution system is applied in an enterprise scenario, business modules such as human resources, sales, and marketing can be abstracted into separate multi-agent systems. Examples include a human resources multi-agent system, a sales multi-agent system, and a marketing multi-agent system.

[0060] The storage section stores information and memories generated during the operation of each multi-agent system. Information can include the capabilities and resources of the multi-agent system. Memories can be understood as log records generated by the multi-agent system during task execution using a large artificial intelligence model.

[0061] The resources section provides the databases, application programming interfaces (APIs), gateways, and other resources required by each multi-agent system. A multi-agent system can have its own database, storing its own capability information within that database.

[0062] The tools section provides the execution tools that the intelligent agent needs to call when performing task steps. These include application programming interfaces (APIs), mail senders, encoders, workflows, and image processors.

[0063] The foundational component provides a subscription platform, along with management functions such as permissions and auditing. The subscription platform supports subscriptions between different multi-agent systems, as well as subscriptions from agents within a multi-agent system to execution tools. Both multi-agent systems and execution tools can be registered on the subscription platform. A multi-agent system can publish its capability information on the subscription platform for invocation. The subscription platform stores the capability information of each multi-agent system. Permissions support the control of user or process access by multi-agent systems, preventing unauthorized access or malicious tampering. Auditing records, monitors, and analyzes relevant behaviors during the operation of multi-agent systems, enabling abnormal behavior detection.

[0064] Based on this, each multi-agent system in the business execution system can register with the subscription platform and subscribe to other multi-agent systems according to actual business dependencies to achieve interaction between business modules. For example, Figure 2 The sales multi-agent system can subscribe to human resources multi-agent systems and marketing multi-agent systems, etc. For example, Figure 2 The market multi-agent system can be subscribed to and sold through multi-agent systems, etc.

[0065] Thus, the first planning agent 101 can obtain second capability information from the subscription platform, namely the capabilities of each second execution agent 201 within the second multi-agent system 200.

[0066] Furthermore, each multi-agent system can have its own independent memory space, and different multi-agent systems can share the memory of related services during cross-multi-agent system calls. Thus, the first planning agent 101 can also obtain second memory information from the subscription platform. The second memory information represents the record information generated by the second multi-agent system 200 when executing services related to the aforementioned service requests, supporting the first planning agent 101 in responding to service requests more accurately.

[0067] Furthermore, the various execution tools included in the business execution system can be registered on the subscription platform. Execution tools can be categorized into public tools and private tools. Public tools are execution tools that are accessible to all agents in the multi-agent system. Private tools are execution tools that are accessible to specific agents in the multi-agent system. For example, execution tools have corresponding access rules set to indicate which execution agents are allowed to access them. Access rules for public tools can allow access from all agents in the multi-agent system. Access rules for private tools can allow access from specific agents in the multi-agent system.

[0068] Furthermore, the execution agent can be used to invoke execution tools from the subscription platform to perform task steps, thereby extending the capabilities of the execution agent. For example, when assigning a first planning agent 101 to execute a task corresponding to a business request, the first planning agent 101 can invoke an authorized execution tool from the subscription platform to execute the corresponding task steps. Based on this, the agent can use one or more published tools to extend its capabilities.

[0069] In this way, the multi-agent system can execute specific tasks by calling the underlying execution tools. If new functions are needed to meet business requirements, new multi-agent systems or execution tools can be deployed in the business execution system. Therefore, the business execution system has high scalability and flexible collaboration capabilities, and can effectively meet the business needs of various complex business scenarios.

[0070] In one possible implementation, combining Figure 1 and Figure 2 The business execution system shown is as follows: Figure 3 The diagram shown is a structural schematic of another business execution system provided in an embodiment of this application.

[0071] Figure 3The agents in the illustrated business execution system can be divided into user agents, planning agents, and execution agents. Furthermore, the business execution system can also include multiple agent pipelines, such as email sending pipelines and work order creation pipelines. An agent pipeline can include multiple agents executing sequentially, which can be used to instruct the execution order of multiple task steps by multiple agents. Different agent pipelines can correspond to different types of business requests. For example, staff can pre-program one or more agent pipelines for at least one multi-agent system and set the execution flow of multiple agents in each agent pipeline, to further publish the multiple agent pipelines on a subscription platform. The subscription platform can be used to store multiple agent pipelines. In this way, the business execution system can support the visualization of complex business processes as individual task steps in agent pipelines, improving the predictability of task execution and effectively supporting the accurate execution of tasks.

[0072] further, Figure 3 In the illustrated business execution system, the multi-agent system, agent pipeline, agents, and execution tools are all independently subscribed to. Thus, the multi-agent system can obtain the necessary agent pipeline, agents, and execution tools from the subscription platform. The agent pipeline can orchestrate agents and execution tools through the subscription platform. Agents can invoke execution tools through the subscription platform. Based on this, when executing a business request, a multi-agent system can invoke agents in other multi-agent systems to implement the corresponding functions, thereby expanding its capabilities. In this way, the business execution system can improve task execution efficiency and system scalability through this multi-level subscription model.

[0073] Based on this, the first planning agent 101 can obtain the target agent pipeline corresponding to the business request from the subscription platform. For example, the first planning agent 101 can identify the agent pipeline of the type corresponding to the business request from among the multiple agent pipelines on the subscription platform as the target agent pipeline. Furthermore, the first planning agent 101 can determine the target execution agent that meets the capability requirements indicated by the target agent pipeline from the first execution agent and / or the second execution agent based on the first capability information and the second capability information, and further schedule the target execution agent to execute the task corresponding to the business request according to the execution order of at least one task step indicated by the target agent pipeline.

[0074] Figure 3 The illustrated business execution system also includes a user agent. The user agent supports interaction between the multi-agent system and the user device. For example, the user agent can receive business requests from the user device and send these requests to the planning agent, thereby processing the business requests. Alternatively, the user agent can receive the processing results of other agents on business requests and return these results to the user device.

[0075] Based on this, the first multi-agent system 100 may include a first user intelligent agent. The first user intelligent agent can be used to receive service requests from user equipment. After the target execution intelligent agent executes the task corresponding to the service request, it can return the execution result to the first planning intelligent agent 101. Furthermore, the first planning intelligent agent 101 can send the execution result corresponding to the service request to the first user intelligent agent. The first user intelligent agent can return the execution result corresponding to the service request to the user equipment.

[0076] Thus, the business execution system in this embodiment can support the independent operation of each business block in the business through multiple multi-agent systems in a distributed structure, as well as the interaction between different business blocks through different multi-agent systems, exhibiting high flexibility and collaboration capabilities. Furthermore, it can support the addition and deployment of new multi-agent systems or execution tools to the business execution system to meet diverse business needs, demonstrating high scalability. Therefore, the business execution system can effectively meet the business needs of various complex business scenarios, avoiding the performance bottlenecks encountered by a single framework when handling large-scale concurrent requests.

[0077] In terms of hardware implementation, the computing device that deploys the aforementioned multi-agent system (such as the first multi-agent system 100 or the second multi-agent system 200) can be implemented via, for example... Figure 4 The structure shown implements the corresponding function. For example... Figure 4 The diagram shown is a hardware structure schematic of a computing device 30 provided in an embodiment of this application.

[0078] Figure 4 The computing device 30 shown may include a processor 301, a memory 302, a communication interface 303, and a bus 304. The processor 301, the memory 302, and the communication interface 303 can be connected via the bus 304.

[0079] The processor 301 is the control center of the computing device 30, and can be a CPU or other general-purpose processor. The general-purpose processor can be a microprocessor or any conventional processor.

[0080] As an example, processor 301 may include one or more CPUs, for example Figure 4 CPU 0 and CPU 1 are shown in the diagram.

[0081] The memory 302 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0082] In one possible implementation, the memory 302 can exist independently of the processor 301. The memory 302 can be connected to the processor 301 via a bus 304 and is used to store data, instructions, or program code. When the processor 301 calls and executes the instructions or program code stored in the memory 302, it can implement the business execution method provided in the embodiments of this application.

[0083] In another possible implementation, the memory 302 can also be integrated with the processor 301.

[0084] The communication interface 303 is used for connecting the computing device 30 to other devices via a communication network, which can be Ethernet, radio access network (RAN), wireless local area network (WLAN), etc. The communication interface 303 may include a receiving unit for receiving data and a transmitting unit for transmitting data.

[0085] Bus 304 can be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, a high-speed peripheral component interconnect (PCI-Express or PCIE) bus, or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0086] It should be pointed out that, Figure 4 The structure shown does not constitute a limitation on computing device 30, except... Figure 4 In addition to the components shown, computing device 30 may include more or fewer components than illustrated, or combine certain components, or have different component arrangements.

[0087] like Figure 5 The diagram shown is a flowchart illustrating a business execution method provided in an embodiment of this application. Figure 5 The method shown can be applied to the aforementioned computing device. The aforementioned [device / equipment] can be deployed within the computing device. Figures 1 to 3 The method includes any one or more business execution systems shown in the figure. The business execution method includes: S401-S402.

[0088] S401. In response to a received business request, invoke one or more multi-agent systems to process the business request.

[0089] Each multi-agent system comprises one or more intelligent agents. An intelligent agent is an entity with independent computing and decision-making capabilities, capable of operating autonomously and interacting with other intelligent agents.

[0090] At least one multi-agent system has an agent pipeline. An agent pipeline consists of multiple agents executing sequentially. Different agent pipelines can be used to handle different types of business logic.

[0091] A business request can be a request to process a specified object. For example, a business request could be a production request for a specified product in a smart factory. Or, a business request could be a delivery request for a specified package in logistics and distribution.

[0092] In one possible implementation, the computer device can receive service requests from user devices, parse the service requests to obtain service information, and then invoke one or more multi-agent systems to process the service requests based on the service information. Specifically, the computer device can read the capability information of each multi-agent system and invoke one or more multi-agent systems whose capability information matches the service information to process the service requests.

[0093] S402. During the process of handling business requests, the intelligent agent pipeline is invoked.

[0094] When multiple agents are called upon to collaboratively process business requests, generating separate tasks for each agent can easily lead to multiple tasks corresponding to the same business request, increasing the probability of errors during processing. To effectively reduce the number of tasks and lower the probability of errors during agent task execution, computer devices can utilize agent pipelines during business request processing. This facilitates the processing of business requests by multiple agents, improving the accuracy of task orchestration and execution efficiency.

[0095] In one possible implementation, in the above S401, when the computer device responds to the received service request and calls one or more multi-agent systems to process the service request, this application embodiment provides an optional implementation including: S4011-S4012.

[0096] S4011. In response to a business request, determine the business information corresponding to the business request.

[0097] The business information corresponding to a business request refers to the business type, business object, and time limit information requested by the business request.

[0098] S4012. Based on business information, call one or more multi-agent systems to process business requests.

[0099] Based on this, the computer device can parse the business request and obtain the corresponding business information. Then, based on the business information, the computer device can invoke one or more multi-agent systems that match the business request to process it. For example, the computer device can invoke a multi-agent system corresponding to the requested business type to process the business request. Or, based on the time limit information of the business request, the computer device can invoke a multi-agent system that can meet the time limit requirements of the business request to process it.

[0100] In one possible implementation, each multi-agent system includes one or more intelligent agents, among which a user intelligent agent may be configured. The user intelligent agent can be used to receive service requests from user devices and parse the service requests to obtain service information, thereby supporting the interaction between the multi-agent system and the user devices. Based on this, in S4011 above, when the computer device responds to a service request and determines the service information corresponding to the service request, the computer device may respond to the service request by invoking the user intelligent agent to parse the service request and obtain the service information.

[0101] In one possible implementation, in the above S4012, when the computer device calls one or more multi-agent systems to process the business request based on the business information, this application embodiment provides an optional implementation, including: S4012a-S4012b.

[0102] S4012a. Determine the intelligent agent pipeline corresponding to the business type.

[0103] S4012b: Invoke one or more agents in the agent pipeline of the multi-agent system to process business requests in the order indicated by the agent pipeline.

[0104] Based on this, a computer device can determine the agent pipeline corresponding to the business type of a business request from the agent pipelines deployed within one or more multi-agent systems. Then, the computer device can invoke agents in one or more multi-agent systems located on the agent pipelines to process the business request in the order indicated by the agent pipeline, thereby efficiently completing the processing.

[0105] In one possible implementation, the capability information of each multi-agent system in the business execution system can be stored on a subscription platform. The capability information of the multi-agent system may include the agent pipelines deployed within the multi-agent system. Thus, the subscription platform can store multiple agent pipelines. Based on this, in S4012a above, when the computer device determines the agent pipeline corresponding to the business type, the computer device can obtain the agent pipeline corresponding to the business type from the subscription platform. Specifically, the computer device can determine the agent pipeline corresponding to the business type of the business request from the multiple agent pipelines stored on the subscription platform, and read the information of the agent pipeline to facilitate further sequential invocation of the agents on the agent pipeline.

[0106] For example, such as Figure 6 The diagram shown is a flowchart illustrating a business execution process provided in an embodiment of this application. Figure 6 In the first multi-agent system, there are a first user agent, an information analysis agent, a request form creation agent, a work order creation agent, an allocation agent, and a dispatch agent. The second multi-agent system includes an engineer query agent. Furthermore, the agent pipeline includes a request form creation pipeline deployed in the first multi-agent system, and a work order execution pipeline deployed in combination with the second multi-agent system. The request form creation pipeline includes the information analysis agent and the request form creation agent executed sequentially. The work order execution pipeline includes the work order creation agent, the allocation agent, the engineer query agent, and the dispatch agent executed sequentially.

[0107] After a user initiates a service request through their user device, the first user agent in the first multi-agent system receives the service request from the user device and parses it to obtain service information. In this case, the computer device can assign the service request to the request form creation pipeline. In the request form creation pipeline, the information analysis agent can analyze the service information of the service request, and the request form creation agent can further create a request form corresponding to the service information. The request form can be used to indicate the processing requirements corresponding to the service request.

[0108] Furthermore, the computer equipment can invoke the work order execution pipeline to create and execute the work order corresponding to the request order. Specifically, the work order creation agent can create a work order corresponding to the request order. Further, the allocation agent can collaborate with the engineer query agent in the second multi-agent system, enabling the engineer query agent to query engineers and allocate engineers to execute the work order. Then, the dispatch agent can dispatch the work order to the allocated engineer, allowing the allocated engineer to execute the work order.

[0109] In one possible embodiment, multiple execution tools may also be registered on the subscription platform, such as application programming interfaces (APIs), email senders, encoders, workflows, and image processors. One or more execution tools have corresponding access rules set, which are used to indicate which agents have the necessary access permissions. Based on this, the business execution method provided in this application embodiment further includes: S501.

[0110] S501. In the process of processing business requests, at least one execution tool on the subscription platform is invoked through the intelligent agent.

[0111] Intelligent agents can execute corresponding task steps in the business process by invoking at least one execution tool on the subscription platform, achieving efficient processing. Thus, by setting up one or more tools in a multi-agent system, the capabilities of the intelligent agent can be extended. If new functions are needed to meet business requirements, new execution tools can be added and deployed in the business execution system. Therefore, the business execution system possesses high scalability and flexible collaboration capabilities, effectively meeting the business needs of various complex business scenarios.

[0112] In one possible embodiment, when multiple multi-agent systems in the business execution system are deployed on different computer devices, the business execution method provided in this application embodiment further includes: S601.

[0113] S601, Send capability information to the subscription platform.

[0114] Capability information is used to indicate the available agent pipelines and / or execution tools.

[0115] In this way, a computer device can register the capability information of the deployed multi-agent system on the subscription platform. Subsequently, other computer devices can use the subscription platform to invoke the intelligent agent pipeline and / or execution tools on that computer device, thereby enabling collaborative processing between multiple multi-agent systems within the business execution system through the subscription platform.

[0116] Based on this, the business execution system can support the independent operation of each business block within a distributed architecture through multiple multi-agent systems, while also enabling interaction between different business blocks through different multi-agent systems, thus possessing high flexibility and collaboration capabilities. If new functionalities are required to meet business needs, new multi-agent systems or execution tools can be added and deployed to the business execution system, demonstrating high scalability. Therefore, the business execution system in this embodiment can effectively meet the business needs of various complex business scenarios, avoiding the performance bottlenecks encountered by a single framework when handling large-scale concurrent requests.

[0117] The foregoing primarily describes the solutions of the embodiments of this application from a methodological perspective. It is understood that, in order to achieve the aforementioned functions, the first planning agent includes at least one of the hardware structures and software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0118] This application embodiment can divide the first planning agent into functional units according to the above method example. For example, each function can be divided into different functional units, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software functional unit. It should be noted that the unit division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0119] When dividing functional units according to their respective functions, such as Figure 7 The diagram shown is a functional structure diagram of a computing device provided in an embodiment of this application. Figure 7 The computing device 40 shown includes: a calling unit 701.

[0120] Calling unit 701 is used to respond to a received business request by calling one or more multi-agent systems to process the business request. Each multi-agent system includes one or more intelligent agents. At least one multi-agent system is equipped with an intelligent agent pipeline, which includes multiple intelligent agents executed in sequence. Calling unit 701 is also used to call the intelligent agent pipeline during the process of processing the business request.

[0121] In one possible implementation, the calling unit 701 is specifically used to: respond to a business request, determine the business information corresponding to the business request; and, based on the business information, call one or more multi-agent systems to process the business request.

[0122] In one possible implementation, the business information includes the business type; the calling unit 701 is specifically used to: determine the agent pipeline corresponding to the business type; and, in accordance with the order indicated by the agent pipeline, call one or more agents in the agent pipeline of the multi-agent system to process the business request.

[0123] In one possible implementation, the calling unit 701 is specifically used to: obtain the agent pipeline corresponding to the business type from the subscription platform; the subscription platform is used to store multiple agent pipelines.

[0124] In one possible implementation, one or more intelligent agents include a user intelligent agent; the invocation unit 701 is specifically used to: in response to a business request, invoke the user intelligent agent to parse the business request and obtain business information.

[0125] In one possible implementation, the calling unit 701 is also used to call at least one execution tool on the subscription platform through the agent during the processing of business requests.

[0126] In one possible implementation, multiple execution tools are registered on the subscription platform; one or more execution tools are configured with access rules, which are used to indicate which agents have the necessary access rights.

[0127] In one possible implementation, the service execution apparatus further includes a sending unit 702. The sending unit 702 is used to send capability information to the subscription platform; the capability information is used to indicate available intelligent agent pipelines and / or execution tools.

[0128] For a detailed description of the above-mentioned optional methods, please refer to the foregoing method embodiments, which will not be repeated here. Furthermore, the explanation of any of the computing devices 40 provided above and the description of their beneficial effects can be found in the corresponding method embodiments described above, and will not be repeated here.

[0129] This application also provides a computer-readable storage medium storing a computer program that, when run on a computer, causes the computer to perform the methods executed by any of the computing devices described above.

[0130] For explanations of the relevant content and descriptions of the beneficial effects in any of the computer-readable storage media provided above, please refer to the corresponding embodiments described above, which will not be repeated here.

[0131] This application also provides a chip. The chip integrates a control circuit for implementing the functions of the aforementioned computing device 30 or computing device 40, and one or more ports. Optionally, the functions supported by the chip can be referred to above, and will not be repeated here. Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium. The aforementioned storage medium can be a read-only memory, random access memory, etc. The aforementioned processing unit or processor can be a central processing unit, a general-purpose processor, an application-specific integrated circuit (ASIC), a microprocessor (digital signal processor, DSP), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0132] This application also provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform any of the methods described in the above embodiments. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions may be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or may include one or more data storage devices such as servers or data centers that can be integrated with the medium. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0133] It should be noted that the devices for storing computer instructions or computer programs provided in the embodiments of this application, such as but not limited to the memory, computer-readable storage medium and communication chip, are all non-transitory.

[0134] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software programs, implementation can be, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device containing one or more servers, data centers, etc., that can be integrated with the medium. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state disks, SSDs).

[0135] Although this application has been described herein in conjunction with various embodiments, those skilled in the art can understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the appended claims in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple components. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results. Various modifications and combinations can be made without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely exemplary illustrations of the application as defined by the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications of this application fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and modifications.

Claims

1. A business execution method, applied to a computing device, characterized in that, The method includes: In response to a received business request, one or more multi-agent systems are invoked to process the business request. Each multi-agent system includes one or more agents. At least one of the multi-agent systems is equipped with an agent pipeline, which includes multiple agents executed sequentially. During the processing of the business request, the intelligent agent pipeline is invoked.

2. The business execution method according to claim 1, characterized in that, The step of responding to a received business request by invoking one or more multi-agent systems to process the business request includes: In response to the service request, determine the service information corresponding to the service request; Based on the business information, one or more multi-agent systems are invoked to process the business request.

3. The business execution method according to claim 2, characterized in that, The business information includes the business type; the step of invoking one or more multi-agent systems to process the business request based on the business information includes: Determine the intelligent agent pipeline corresponding to the aforementioned business type; The business request is processed by invoking one or more agents in the agent pipeline of the multi-agent system in the order indicated by the agent pipeline.

4. The business execution method according to claim 3, characterized in that, The process of determining the intelligent agent pipeline corresponding to the business type includes: Obtain the agent pipeline corresponding to the business type from the subscription platform; the subscription platform is used to store multiple agent pipelines.

5. The business execution method according to claim 2, characterized in that, The one or more intelligent agents include a user intelligent agent; the step of determining the business information corresponding to the business request in response to the business request includes: In response to the business request, the user intelligent agent is invoked to parse the business request and obtain the business information.

6. The business execution method according to claim 2, characterized in that, The method further includes: In processing the business request, the agent invokes at least one execution tool on the subscription platform.

7. The business execution method according to claim 6, characterized in that, The subscription platform has multiple execution tools registered; one or more execution tools have access rules set for them, and the access rules are used to indicate which intelligent agents have access rights.

8. The business execution method according to claim 1, characterized in that, The method further includes: Send capability information to the subscription platform; the capability information is used to indicate the intelligent agent pipeline and / or execution tools that can be invoked.

9. A business execution system, the business execution system being used to run on one or more computing devices, characterized in that, include: One or more multi-agent systems, the multi-agent systems being used to process business requests; Each of the multi-agent systems includes one or more agents, and at least one of the multi-agent systems is equipped with an agent pipeline, which includes multiple agents executed sequentially. The multi-agent system is used to invoke the agent pipeline during the processing of the business request.

10. A computing device, characterized in that, The computing device is used to execute the business execution method as described in any one of claims 1 to 8, or the computing device is equipped with the business execution system as described in claim 9.

Images