Intelligent scheduling method, system, device, equipment and medium

Abstract

The invention discloses an intelligent scheduling method, system and device, equipment and a medium. The method is applied to a core agent in an intelligent scheduling system, and comprises the following steps: acquiring target interaction information input by a target user; generating a target work plan according to the target interaction information; the target work plan comprises at least one target task and at least one type of target execution information such as a target task type and a target execution module identifier corresponding to the target task; calling a corresponding target agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task to obtain a corresponding target execution result; the target agent module comprises at least one of the following: a single agent, an agent chain formed by connecting a plurality of agents in series, and an agent group formed by a plurality of agent chains; the target interaction result of the target interaction information is determined according to the target execution result, the overall completion efficiency of complex tasks such as software development can be improved, and the intelligence of cooperative work among multiple agents can be improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an intelligent scheduling method, system, device, equipment and medium. Background Art

[0002] In the traditional software development process, developers are often required to manually handle complex task processes, including plan generation, demand generation, architecture design, task decomposition, code generation, test case generation, etc. There are dependencies between these tasks, which requires developers to spend a lot of time and energy to coordinate and manage. Summary of the invention

[0003] The embodiments of the present application provide an intelligent scheduling method, system, device, equipment and medium, which realizes the use of collaborative work between multiple agents to handle relatively complex task processes, saving the time and energy required by developers or users who need to complete auxiliary tasks in coordinating and managing task processes, and can improve the overall completion efficiency of complex tasks such as software development, and can improve the intelligence of collaborative work between multiple agents. The above technical solutions are as follows: In a first aspect, an embodiment of the present application provides an intelligent scheduling method, which is applied to a core agent in an intelligent scheduling system; the method includes: Obtain target interaction information input by the target user; Generate a target work plan according to the target interaction information; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: the target task type corresponding to the target task and the target execution module identifier; According to the target execution information corresponding to the target task, the corresponding target agent module in the intelligent scheduling system is called to execute the target task to obtain the corresponding target execution result; the target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; The target interaction result corresponding to the target interaction information is determined according to the target execution result.

[0004] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes target execution orders corresponding to the multiple target tasks. The above-mentioned generating a target work plan according to the above-mentioned target interaction information includes: Analyze the above target interaction information to obtain the total target interaction task; Decomposing the above target interaction total task to obtain the above multiple target tasks; Determining the target execution order corresponding to each of the plurality of target tasks according to the target task types corresponding to each of the plurality of target tasks; According to the target execution information corresponding to the target task, the corresponding target agent module in the intelligent scheduling system is called to execute the target task, and the corresponding target execution result is obtained, including: According to the target task types and / or target execution module identifiers corresponding to the above-mentioned multiple target tasks, the corresponding target agent modules in the above-mentioned intelligent scheduling system are called in sequence in the above-mentioned target execution order to execute the above-mentioned target tasks, so as to obtain the target execution results corresponding to the above-mentioned multiple target tasks.

[0005] In a possible implementation, after the target interaction overall task is decomposed to obtain the multiple target tasks, the method further includes: According to the target task type corresponding to the above-mentioned target task and the functions and / or status information corresponding to each intelligent agent module in the above-mentioned intelligent scheduling system, the target intelligent agent module that executes the above-mentioned target task is determined; the above-mentioned target execution module identifier is the module identifier corresponding to the above-mentioned target intelligent agent module.

[0006] In a possible implementation, the target work plan is generated according to the target interaction information, including: analyzing the target interaction information to obtain a target interaction overall task; determining the target work plan according to a preset plan rule and a target type corresponding to the target interaction overall task; different target types correspond to different target work plans.

[0007] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution type and a dependent execution type; According to the target execution information corresponding to the target task, the corresponding target agent module in the intelligent scheduling system is called to execute the target task, and the corresponding target execution result is obtained, including: According to the target task type and / or target execution module identifier corresponding to the first target task, the first target task is assigned to the corresponding first target agent module in the above-mentioned intelligent scheduling system, and the first target execution result output by the above-mentioned first target agent module is obtained; the above-mentioned first target task is a target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; In the case where there is a second target task among the plurality of target tasks, target input information corresponding to the second target task is generated according to the execution result of the first target on which the second target task depends; the second target task is a target task belonging to the dependent execution class; According to the target task type and / or target execution module identifier corresponding to the above-mentioned second target task, the above-mentioned second target task and the above-mentioned target input information are sent to the corresponding second target intelligent agent module in the above-mentioned intelligent scheduling system, so that the above-mentioned second target intelligent agent module executes the above-mentioned second target task according to the above-mentioned target input information, and obtains the second target execution result output by the above-mentioned second target intelligent agent module.

[0008] In one possible implementation, the intelligent scheduling system includes multiple intelligent agent modules, and the database of the core intelligent agent stores at least one of the following shared information: task execution information corresponding to each of the multiple intelligent agent modules, system information corresponding to the intelligent scheduling system, and task auxiliary information.

[0009] In a possible implementation, the method further includes: When the amount of information corresponding to the shared information is greater than a threshold, extracting key information and overview information from the shared information; Update the above shared information based on the above key information and the above overview information.

[0010] In a possible implementation, after calling the corresponding target agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task, the method further includes: The target execution status and target execution result corresponding to the above target work plan are displayed according to the current execution status of the above target intelligent agent module.

[0011] In a possible implementation, the target agent modules are different or the types corresponding to the target agent modules are different, and the target functions corresponding to the target agent modules are different; The above-mentioned target tasks include at least one of the following: requirement generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks and question and answer tasks.

[0012] In a second aspect, an embodiment of the present application provides another intelligent scheduling method, which is applied to a target agent module in an intelligent scheduling system; the target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; the method includes: Receiving a target task assigned by a core agent in the intelligent scheduling system according to a target work plan; the target work plan is generated by the core agent according to target interaction information input by a target user; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: a target task type corresponding to the target task and a target execution module identifier; Execute the above target tasks and obtain the corresponding target execution results; The target execution result is sent to the core agent, so that the core agent determines the target interaction result corresponding to the target interaction information according to the target execution result.

[0013] In a third aspect, an embodiment of the present application provides an intelligent scheduling system, the intelligent scheduling system comprising a core agent and multiple agent modules; the agent modules comprising at least one of the following: a single agent, an agent chain consisting of multiple agents connected in series, and an agent group consisting of multiple of the above agent chains; The core agent is used to obtain the target interaction information input by the target user, and generate a target work plan according to the target interaction information, wherein the target work plan includes at least one target task and target execution information corresponding to the at least one target task; and, according to the target work plan, call the corresponding target agent module in the intelligent scheduling system to execute the target task and obtain the corresponding target execution result; and, according to the target execution result, determine the target interaction result corresponding to the target interaction information; The above-mentioned intelligent agent module is used to execute the tasks assigned by the above-mentioned core intelligent agent.

[0014] In a possible implementation, at least one of the following shared information is shared and stored in the database of the core agent: task execution information corresponding to each of the multiple agent modules, system information corresponding to the intelligent scheduling system, and task auxiliary information.

[0015] In a possible implementation, the above-mentioned intelligent agent modules are different or the types corresponding to the above-mentioned intelligent agent modules are different, and the above-mentioned intelligent agent modules correspond to different target functions; The above-mentioned target tasks include at least one of the following: requirement generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks and question and answer tasks.

[0016] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution type and a dependent execution type; After the first target intelligent agent module corresponding to the first target task among the above-mentioned multiple target tasks obtains the corresponding first target execution result, the above-mentioned first target execution result is sent to the second target intelligent agent module corresponding to the dependent second target task, so as to trigger the above-mentioned second target intelligent agent module to execute the above-mentioned second target task according to the first target execution result that the above-mentioned second target task depends on, and obtain the second target execution result corresponding to the second target task; wherein the above-mentioned first target task is the target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; the above-mentioned second target task is the target task belonging to the above-mentioned dependent execution class among the above-mentioned multiple target tasks.

[0017] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution type and a dependent execution type; After the first target agent module corresponding to the first target task among the above-mentioned multiple target tasks obtains the corresponding first target execution result, the first target execution result is sent to the above-mentioned core agent; the above-mentioned first target task is a target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; In the case where there is a second target task among the above-mentioned multiple target tasks, the above-mentioned core agent generates target input information corresponding to the above-mentioned second target task according to the execution result of the first target on which the above-mentioned second target task depends, and sends the above-mentioned second target task and the above-mentioned target input information to the corresponding second target agent module; the above-mentioned second target task is a target task belonging to the above-mentioned dependent execution class among the above-mentioned multiple target tasks; The second target agent module executes the second target task according to the target input information to obtain the second target execution result.

[0018] In a fourth aspect, an embodiment of the present application provides an intelligent scheduling device, which is applied to a core intelligent entity in an intelligent scheduling system; the intelligent scheduling device includes: An acquisition module, used to acquire target interaction information input by a target user; A plan generation module, configured to generate a target work plan according to the target interaction information; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: a target task type corresponding to the target task and a target execution module identifier; A calling module is used to call the corresponding target agent module in the above-mentioned intelligent scheduling system to execute the above-mentioned target task according to the target execution information corresponding to the above-mentioned target task, and obtain the corresponding target execution result; the above-mentioned target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents connected in series, and an agent group composed of multiple of the above-mentioned agent chains; A determination module is used to determine a target interaction result corresponding to the target interaction information according to the target execution result.

[0019] In a fifth aspect, an embodiment of the present application provides another intelligent scheduling device, which is applied to a target agent module in an intelligent scheduling system; the target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; the intelligent scheduling device includes: A receiving module is used to receive a target task assigned by a core agent in the intelligent scheduling system according to a target work plan; the target work plan is generated by the core agent according to target interaction information input by a target user; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: a target task type corresponding to the target task and a target execution module identifier; An execution module is used to execute the above target tasks and obtain corresponding target execution results; The sending module is used to send the above-mentioned target execution result to the above-mentioned core intelligent agent, so that the above-mentioned core intelligent agent determines the target interaction result corresponding to the above-mentioned target interaction information according to the above-mentioned target execution result.

[0020] In a sixth aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory; The processor is connected to the memory; The above-mentioned memory is used to store executable program code; The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the first aspect of the embodiment of this specification or any possible implementation of the first aspect or the method provided by the second aspect.

[0021] In the seventh aspect, an embodiment of the present specification provides a computer storage medium, wherein the computer storage medium stores multiple instructions, and the instructions are suitable for being loaded by a processor and executing the first aspect of the embodiment of the present specification or any possible implementation of the first aspect or the method provided by the second aspect.

[0022] In one or more embodiments of the present application, the core intelligent agent in the intelligent scheduling system can be used to interact with the target user, and the corresponding target work plan can be generated based on the target interaction information input by the target user, and the appropriate target intelligent agent module can be selected to call and execute each target task in the target work plan based on the target execution information corresponding to each target task in the target work plan, and the target interaction result corresponding to the target interaction information can be determined based on the target execution result corresponding to each target task, thereby realizing the use of collaborative work between multiple agents to handle relatively complex task processes, saving the time and energy required by developers or users who need to complete complex tasks in coordinating and managing task processes, thereby improving the overall completion efficiency of complex tasks such as software development, and improving the intelligence of collaborative work between multiple agents.

[0023] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0025] Figure 1 A schematic diagram of the architecture of an intelligent scheduling system provided for an exemplary embodiment of the present application; Figure 2 A flowchart of an intelligent scheduling method provided by an exemplary embodiment of the present application; Figure 3 A schematic diagram of a process for generating a target work plan provided by an exemplary embodiment of the present application; Figure 4 A schematic diagram of an interactive effect of intelligent scheduling provided by an exemplary embodiment of the present application; Figure 5 A schematic diagram of another interactive effect of intelligent scheduling provided by an exemplary embodiment of the present application; Figure 6 A flowchart of another intelligent scheduling method provided for an exemplary embodiment of the present application; Figure 7 A schematic diagram of an implementation process of an intelligent scheduling method provided by an exemplary embodiment of the present application; Figure 8A schematic diagram of the structure of an intelligent scheduling device provided by an exemplary embodiment of the present application; Fig. 9 A schematic diagram of the structure of another intelligent scheduling device provided by an exemplary embodiment of the present application; Fig.10 A schematic structural diagram of an electronic device provided as an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0027] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0028] Please refer to the following Figure 1 , which is a schematic diagram of the architecture of an intelligent scheduling system provided by an exemplary embodiment of this specification. Figure 1 As shown, the intelligent scheduling system may include: a core intelligent agent 110 and multiple intelligent agent modules. The core intelligent agent 110 is the core of the intelligent scheduling system, which is used to obtain the target interaction information input by the target user, and generate a target work plan based on the target interaction information. The above target work plan includes at least one target task and target execution information corresponding to at least one target task; and, according to the target work plan, call the corresponding target intelligent agent module in the intelligent scheduling system to execute the target task and obtain the corresponding target execution result; and, determine the target interaction result corresponding to the target interaction information based on the target execution result.

[0029] The agent module may include but is not limited to at least one of the following: a single agent 121, an agent chain 122 composed of multiple agents connected in series, and an agent group 123 composed of multiple agent chains. The agent module is used to execute the tasks assigned by the core agent 110.

[0030] It can be understood that an intelligent agent can be a unit that can autonomously perceive the environment, make autonomous decisions and take actions. It can be a physical entity, such as a robot or an aircraft, or an abstract entity, such as a character in a virtual game, etc. The embodiments of the present application are not limited to this.

[0031] In an embodiment of the present application, a core intelligent agent can be set for the intelligent scheduling system, and the core intelligent agent can be used to interact with the target user, and the corresponding target work plan can be considered based on the target interaction information input by the target user, and the appropriate target intelligent agent module can be selected to call and execute the target task in the target work plan, thereby realizing the use of collaborative work between multiple agents to handle relatively complex task processes, saving the time and energy required for developers or users who need to complete tasks to coordinate and manage task processes, and improving the completion efficiency of complex tasks such as software development tasks.

[0032] Optionally, the database of the core agent 110 stores at least one of the following shared information: task execution information corresponding to each of the multiple agent modules (such as, but not limited to, task execution logs corresponding to each agent module, such as task name, task input, task output, task status, etc.), system information corresponding to the intelligent scheduling system (such as, but not limited to, storage system task execution time, role name description or function type description corresponding to each agent module, etc.), and task auxiliary information (such as, but not limited to, knowledge graph information corresponding to the fields of various functions of the intelligent scheduling system, knowledge information obtained by the core agent 110 or each agent module in the process of calling other tools to perform tasks, etc.). The above-mentioned shared information can be shared between the core agent 110 and multiple agent modules in the intelligent scheduling system to assist the core agent and each agent module in completing the tasks they need to perform.

[0033] Optionally, the above-mentioned intelligent agent modules are different or the types corresponding to the intelligent agent modules are different, and the target functions corresponding to the above-mentioned intelligent agent modules are different, and the target tasks that can be performed are also different. The above-mentioned target tasks may include but are not limited to at least one of the following: demand generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks, and question-answering tasks.

[0034] That is, in the embodiment of the present application, the core intelligent agent 110 in the intelligent scheduling system can call multiple intelligent agents to work together to implement the entire software development process according to user needs, thereby improving software development efficiency; and it can also provide users with question-and-answer services related to software development projects, efficiently and accurately solving various problems encountered by users in the software development process.

[0035] Optionally, the above-mentioned target work plan may include, but is not limited to, multiple target tasks, and the target execution information may also include, but is not limited to, the target execution type corresponding to the target task, and the above-mentioned target execution type may include, but is not limited to, at least one of the following: an independent execution class and a dependent execution class. After the first target intelligent agent module corresponding to the first target task among the above-mentioned multiple target tasks obtains the corresponding first target execution result, the above-mentioned first target execution result may be directly sent to the second target intelligent agent module corresponding to the second target task that is dependent, so as to trigger the above-mentioned second target intelligent agent module to execute the above-mentioned second target task according to the first target execution result that the above-mentioned second target task depends on, and obtain the second target execution result corresponding to the second target task; the above-mentioned first target task is a target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; the above-mentioned second target task is a target task belonging to the above-mentioned dependent execution class among the above-mentioned multiple target tasks.

[0036] That is, in the embodiments of the present application, the intelligent agent module in the intelligent scheduling system can not only communicate with the core intelligent agent, such as receiving the target tasks assigned by the core intelligent agent, sending the corresponding target execution results to the core intelligent agent, etc., but also communicate with other intelligent agent modules, such as directly transmitting the execution results on which its execution tasks depend to other intelligent agent modules, etc., to avoid the problem of a large number of execution results being repeatedly copied, pasted and forwarded, wasting system resources, and to a certain extent improve the execution efficiency of complex tasks and reduce the consumption of communication resources for executing complex tasks.

[0037] Optionally, the above-mentioned target work plan may include, but is not limited to, multiple target tasks, and the above-mentioned target execution information also includes the target execution type corresponding to the above-mentioned target task; the above-mentioned target execution type includes at least one of the following: independent execution class and dependent execution class. After the first target intelligent agent module corresponding to the first target task among the above-mentioned multiple target tasks obtains the corresponding first target execution result, the above-mentioned first target execution result is sent to the above-mentioned core intelligent agent 110; the above-mentioned first target task is the target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; in the case that there is a second target task among the above-mentioned multiple target tasks, the above-mentioned core intelligent agent 110 generates the target input information corresponding to the above-mentioned second target task according to the first target execution result on which the above-mentioned second target task depends, and sends the above-mentioned second target task and the above-mentioned target input information to the corresponding second target intelligent agent module; the above-mentioned second target task is the target task belonging to the above-mentioned dependent execution class among the above-mentioned multiple target tasks; the above-mentioned second target intelligent agent module executes the above-mentioned second target task according to the above-mentioned target input information to obtain the second target execution result.

[0038] That is, in the embodiments of the present application, the core intelligent agent in the intelligent scheduling system can be used to realize rapid calling and information interaction between multiple intelligent agent modules, such as converting the output (first target execution result) of the previous intelligent agent module (first target intelligent agent module) into the input (target input information) required for the next intelligent agent module (second target intelligent agent module) to perform the task, etc., to facilitate the management of the execution progress of complex tasks in the intelligent scheduling system and the coordination and cooperation between the intelligent agent modules.

[0039] The network may be a medium that provides a communication link between the core agent 110 and each agent module and between multiple agent modules, or may be the Internet including network devices and transmission media, but is not limited thereto. The transmission media may be a wired link, such as but not limited to coaxial cable, optical fiber, and digital subscriber line (DSL), or a wireless link, such as but not limited to wireless fidelity (WIFI), Hypertext Transfer Protocol (HTTP), Bluetooth, and mobile device network, etc.

[0040] Understandably, Figure 1 The number of core agents 110 and agent modules in the intelligent scheduling system shown is only an example. In a specific implementation, the intelligent scheduling system may include any number of core agents 110 and agent modules.

[0041] Next, combine Figure 1 , taking the core agent 110 in the intelligent scheduling system performing intelligent scheduling as an example, an intelligent scheduling method provided by an exemplary embodiment of the present application is introduced. Figure 2 , which exemplarily shows a flow chart of an intelligent scheduling method provided by an embodiment of the present application. Figure 2 As shown, the intelligent scheduling method includes the following steps: S201, obtaining target interaction information input by a target user.

[0042] Specifically, when the target user has task requirements or has some questions that require the intelligent scheduling system to help answer, the corresponding target interaction information can be sent directly to the core intelligent body of the intelligent scheduling system through its terminal. For example, but not limited to, intelligent interaction software can be installed on the terminal of the target user, and the software is set with an intelligent scheduling system to provide corresponding target interaction functions according to the target interaction information input by the target user to meet the various interaction needs of the target user. The above-mentioned target interaction information can be voice interaction information, text interaction information, or image interaction information, which is not limited to the embodiments of the present application. The above-mentioned target interaction information may include, but is not limited to, target interaction demand description information of the target user, such as, but not limited to, target problem description information, target task demand description information, etc.

[0043] S202: Generate a target work plan according to the target interaction information, where the target work plan includes at least one target task and target execution information corresponding to the at least one target task.

[0044] Specifically, the above-mentioned target execution information may include but is not limited to at least one of the following: the target task type corresponding to the target task, and the target execution module identifier. The above-mentioned target task type is used to distinguish the different natures, purposes or execution methods of the target tasks, such as but not limited to demand generation class, architecture generation class, task decomposition class, code generation class, test case generation class and question and answer class. The above-mentioned target execution module identifier may be but is not limited to the module identifier corresponding to the target intelligent agent module that needs to execute the target task, such as but not limited to the target role name or target number. The target intelligent agent modules corresponding to different target execution module identifiers can perform different functions or target task types.

[0045] The above-mentioned target role names reflect the specific applications and functions of the target intelligent agent module in different fields and scenarios. For the same target intelligent agent module, the fields or scenarios corresponding to the target tasks are different, and the target role names corresponding to the target intelligent agent module may also be different. That is, a target intelligent agent module may correspond to a target role name in one field or scenario, or may correspond to multiple target role names in multiple different fields or scenarios. For example, when the target interaction information belongs to information in an educational scenario, the target role name corresponding to the target intelligent agent module may be a math teacher. When the target interaction information belongs to information in a software development scenario, the target role name corresponding to the target intelligent agent module may also be a product manager, etc. The embodiments of the present application do not limit this.

[0046] Optionally, the target work plan includes multiple target tasks, and the target execution information also includes the target execution order corresponding to each of the multiple target tasks. Figure 3As shown, the above S202, the implementation process of generating a target work plan according to the target interaction information may include but is not limited to: first analyzing the target interaction information, and obtaining the target interaction total task according to the analyzed target interaction requirements; then, disassembling the target interaction total task, for example, but not limited to disassembling the target interaction total task according to the number or type of target interaction requirements, to obtain multiple target tasks; finally, according to the target task types corresponding to the multiple target tasks or the association relationship between the multiple target tasks, determining the target execution order corresponding to the multiple target tasks. For multiple target tasks, the target task types are different, and the target execution order of the corresponding target tasks may be different. For example, the target task of the demand generation class needs to be executed before the target task of the architecture generation class.

[0047] Furthermore, after the core agent decomposes the target interaction task and obtains multiple target tasks, it can also determine the target agent module that executes each target task according to the target task type corresponding to each target task and the function and / or status information corresponding to each agent module in the intelligent scheduling system. For example, but not limited to, an agent module that can execute the function corresponding to the target task type and is operating normally or has no other tasks currently being executed is determined as the target agent module that executes the target task. The above-mentioned status information may include, but is not limited to, the current load status information or operation status information of the agent module. The above-mentioned load status information reflects the workload currently being processed by the agent module and the resource consumption required to process these tasks. The above-mentioned operation status information may include, but is not limited to, information such as the health status, configuration status, and working status of key components of the agent module, which is used to identify whether the agent module can operate normally.

[0048] Optionally, after obtaining the target interaction information, the target interaction information can be analyzed first to obtain the target interaction total task; then, according to the preset planning rules, according to the target type corresponding to the target interaction total task, the target work plan is determined; the above target types are different, and the target work plans corresponding to the above target interaction total tasks are different. The above preset planning rules may include a preset correspondence between task types and work plans, for example, but not limited to, the target work plan corresponding to the question-and-answer type target interaction total task is: execute the target question-and-answer task, the target work plan corresponding to the architecture generation type target interaction total task is: first execute the target product demand generation task, then execute the target architecture generation task, etc., or the target interaction information can be directly input into a pre-trained target large model, and the corresponding target work plan is output, etc., and the embodiments of the present application are not limited to this.

[0049] Please continue to refer to Figure 2 ,like Figure 2As shown, in the above S202, after generating the target work plan according to the target interaction information, the intelligent scheduling method may further include: S203, according to the target execution information corresponding to the target task, calling the corresponding target agent module in the intelligent scheduling system to execute the target task and obtain the corresponding target execution result.

[0050] Specifically, the above-mentioned target agent module may include, but is not limited to, at least one of the following: a single agent, an agent chain consisting of multiple agents connected in series, and an agent group consisting of multiple agent chains. When the target agent module is an agent chain, the core agent can call the entire agent chain to perform the target task, or call one or more target agents in the agent chain to perform the target task. The specific selection can be made according to the target task and the actual situation of each agent in the agent chain, and the embodiments of the present application do not limit this. When the target agent module is an agent group, the core agent can call the entire agent group to perform the target task, or call one or more target agent chains in the agent group to perform the target task, or call one or more target agents in the agent group to perform the target task. The specific selection can be made according to the target task and the actual situation of each intelligent chain and agent in the agent group, and the embodiments of the present application do not limit this.

[0051] Optionally, the target execution information includes a target task type corresponding to the target task. After the target work plan is generated, the target intelligent agent module matching the target task type in the intelligent scheduling system can be called to execute the target task according to the target work plan and the target task type corresponding to the target task, for example, but not limited to, calling a target intelligent agent module whose executable function in the intelligent scheduling system matches the target function to be achieved corresponding to the target task type.

[0052] Optionally, the target execution information includes a target execution module identifier corresponding to the target task. After generating the target work plan, the target agent module corresponding to the target execution module identifier in the intelligent scheduling system can be directly called to execute the target task according to the target execution module identifier corresponding to each target task in the target work plan.

[0053] Optionally, the above target execution information may also include, but is not limited to, the target priority (target execution order), target urgency, and required target execution resources corresponding to the target task. The core agent may use built-in artificial intelligence algorithms (such as, but not limited to, genetic algorithms, simulated annealing algorithms, ant colony algorithms, etc.) to intelligently match target tasks with task execution resources (agent modules) based on the target task type, target priority (target execution order), target urgency, and required target execution resources of each target task, and assign the target task to the most appropriate target agent module in the intelligent scheduling system based on the matching results.

[0054] Specifically, after the target agent module completes the target task, it will output the corresponding target execution result. The core agent can obtain and save the target execution results output by each target agent module.

[0055] Optionally, the target work plan includes multiple target tasks, and the target execution information also includes the target execution order corresponding to each of the multiple target tasks. The above S203, based on the target execution information corresponding to the target task, calling the corresponding target intelligent agent module in the intelligent scheduling system to execute the target task, and obtaining the corresponding target execution result may include but is not limited to: based on the target task type and / or target execution module identifier corresponding to each of the multiple target tasks, calling the corresponding target intelligent agent module in the intelligent scheduling system in the target execution order to execute the target task, and obtaining the target execution results corresponding to each of the multiple target tasks.

[0056] Optionally, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type may include but is not limited to at least one of the following: an independent execution class and a dependent execution class. The target task of the independent execution class is a task that can be executed independently and does not need to rely on the execution results of other tasks, and the target task of the dependent execution class is a task that cannot be executed independently and needs to rely on the execution results of other tasks to be executed. The above S203, according to the target execution information corresponding to the target task, calls the corresponding target agent module in the intelligent scheduling system to execute the target task, and obtains the corresponding target execution result, which may include but is not limited to: according to the target task type and / or target execution module identifier corresponding to the first target task, assigns the first target task to the corresponding first target agent module in the intelligent scheduling system, and obtains the first target execution result output by the first target agent module, and the above first target task is a target task belonging to the independent execution class among multiple target tasks; in the case where there is a second target task among multiple target tasks, according to the first target execution result on which the second target task depends, generates the target input information corresponding to the second target task, and according to the target task type and / or target execution module identifier corresponding to the second target task, sends the second target task and the target input information to the corresponding second target agent module in the intelligent scheduling system, so that the second target agent module executes the second target task according to the target input information, and obtains the second target execution result output by the second target agent module. The above second target task is a target task belonging to the dependent execution class. The above target input information may include but is not limited to the first target execution result on which the second target task depends and the target description information corresponding to the second target task.

[0057] It can be understood that the target execution type of each target task in the target work plan may be different if the target interaction information is different. For example, when the target interaction information includes target product requirement information and architecture generation requirement description information, the target task in the target work plan corresponding to the target interaction information only includes independent execution type architecture generation tasks; when the target interaction information only includes architecture generation type requirement description information, the target task in the target work plan corresponding to the target interaction information includes independent execution type product requirement generation tasks and dependent execution type architecture generation tasks of the target product requirements (target execution results) generated after the dependent product requirement generation tasks are executed.

[0058] Please continue to refer to Figure 2 ,like Figure 2 As shown, in the above S203, according to the target execution information corresponding to the target task, the corresponding target agent module in the intelligent scheduling system is called to execute the target task, and after obtaining the corresponding target execution result, the intelligent scheduling method further includes: S204: Determine a target interaction result corresponding to the target interaction information according to the target execution result.

[0059] Specifically, after each target task in the target work plan is completed, the corresponding target agent module will return the target execution result to the core agent. The core agent can integrate and analyze the collected target execution results to obtain the target interaction results corresponding to the target interaction information.

[0060] For example, Figure 4 As shown, after the core agent receives the target interaction information 410 "Hello, what functions do you have" input by the target user, it can analyze the target interaction information 410 and determine that the target task type 420 corresponding to the corresponding target task is "other", and the target task description information 430 is "the most suitable classification is 'other', which means it is responsible for daily conversations and solving various user problems, including answering questions about functions". The target execution module identifier 440 of the target task corresponding to the target interaction information 410 is "LLM Agent", that is, a target agent module based on a large language model (Large Language Model). When the target agent module completes the above-mentioned target task, the core agent can receive the target execution result fed back by it, and determine and feed back the target interaction result 450 corresponding to the target interaction information 410 to the target user based on the above-mentioned target execution result.

[0061] Optionally, after the core intelligent agent obtains the target execution results corresponding to each internal target task in the target work plan, it can also, but is not limited to, evaluate the quality and efficiency of the target work plan and the completion of each target task based on the quality of the target execution results corresponding to each target task (for example, but not limited to accuracy, effectiveness, relevance to target interaction information, etc.), the allocation time and completion time of each target task, etc., and optimize the subsequent work plan generation, task allocation scheduling strategies, etc. based on the completion status, so as to improve the overall efficiency and performance of the intelligent scheduling system.

[0062] In the embodiments of the present application, the core intelligent agent in the intelligent scheduling system can be used to interact with the target user, and the corresponding target work plan can be generated based on the target interaction information input by the target user, and the appropriate target intelligent agent module can be selected to call and execute each target task in the target work plan based on the target execution information corresponding to each target task in the target work plan, and the target interaction result corresponding to the target interaction information can be determined based on the target execution result corresponding to each target task, thereby realizing the use of collaborative work between multiple agents to handle relatively complex task processes, saving the time and energy required by developers or users who need to complete complex tasks in coordinating and managing task processes, thereby improving the overall completion efficiency of complex tasks such as software development tasks, and improving the intelligence of collaborative work between multiple agents.

[0063] In some possible embodiments, the intelligent scheduling system may include multiple agent modules, and the database of the core agent stores at least one of the following shared information: task execution information corresponding to each of the multiple agent modules (such as, but not limited to, task execution logs corresponding to each agent module, such as task name, task input, task output, task status, etc.), system information corresponding to the intelligent scheduling system (such as, but not limited to, storage system task execution time, role name description or function type description corresponding to each agent module, etc.), and task auxiliary information (such as, but not limited to, knowledge graph information corresponding to the fields of various functions of the intelligent scheduling system, knowledge information obtained by the core agent or each agent module in the process of calling other tools to perform tasks, etc.). The above-mentioned shared information can be shared between the core agent and multiple agent modules in the intelligent scheduling system to assist the core agent and each agent module in completing the tasks they need to perform.

[0064] Optionally, when the amount of information corresponding to the above shared information is greater than a threshold, it means that there is too much information currently stored in the database. In this case, the key information and overview information in the shared information can be extracted first, and then the above shared information can be updated based on the key information and overview information, such as but not limited to directly replacing the shared information with the key information and overview information in the shared information, thereby avoiding the problem of excessive amount of information in the database affecting the performance of the intelligent scheduling system.

[0065] Optionally, the shared information in the above-mentioned database can be compressed and stored, and can also support memory retrieval. For example, the target intelligent agent module that performs the target task can retrieve target information related to the target task it performs from the shared information stored in the database to assist it in better completing the target task.

[0066] In some possible embodiments, after the core intelligent agent calls the corresponding target intelligent agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task, the core intelligent agent can also, but is not limited to, display the target execution status and target execution result corresponding to the target work plan according to the current execution status of the target intelligent agent module. For example, but not limited to, when the target work plan includes three target tasks, the current execution status of the target intelligent agent modules corresponding to two of the three target tasks is the execution completion status, and the current execution status of the target intelligent agent module corresponding to the remaining target task is the execution-in-progress status, then the core intelligent agent module can determine that the target execution status corresponding to the target work plan is the execution-in-progress status, and the current execution progress is two-thirds. The core intelligent agent can display and update the target execution status corresponding to the target work plan and the target execution results of each target task in real time according to the current execution status corresponding to the three target tasks, so that the target user can intuitively understand the execution status of the work plan and enhance the interactive experience of the target user and the intelligent scheduling system.

[0067] In some possible embodiments, the target agent modules are different or the types corresponding to the target agent modules are different, the target functions corresponding to the target agent modules are different, and the target tasks that can be performed are also different. The target tasks may include but are not limited to at least one of the following: demand generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks, and question-answering tasks.

[0068] For example, Figure 5As shown, when the target user inputs the target interaction information 510 "help me generate the login interface architecture design", the core agent can analyze it and obtain the target type 520 of the target interaction task as "architecture design", and generate the corresponding target work plan 530 according to the target type 520 and the target interaction information 510, that is, the target agent module corresponding to the product manager "Product Manager" first performs the product demand generation task, and then the target agent module corresponding to the architect "Architecter" performs the architecture design task. After the core agent calls the target agent module corresponding to the product manager "Product Manager" to perform the product demand generation task and obtains the corresponding target execution result 540, it can call the target agent module corresponding to the architect "Architecter" to perform the architecture design task according to the target execution result 540 and the architecture design task description information 550, and obtain the corresponding target execution result 560. When the target agent module corresponding to the architect "Architecter" performs the architecture design task, it can also first analyze and disassemble the architecture design task, and then determine the architecture design subtasks to be performed by each target agent in the target agent module. At the same time, the target intelligent agent module corresponding to the architect "Architecter" can also call other intelligent agent modules to assist it in executing architectural design subtasks (such as code generation tasks corresponding to data structures and interfaces and code generation tasks corresponding to program call processes), and obtain execution results feedback from other intelligent agent modules (such as code 561 corresponding to data structures and interfaces and code 562 corresponding to program call processes).

[0069] Next, combine Figure 1 , taking the intelligent scheduling performed by the target agent module in the intelligent scheduling system as an example, the intelligent scheduling method provided by an exemplary embodiment of the present application is introduced. Figure 6 , which exemplarily shows a flow chart of another intelligent scheduling method provided by an embodiment of the present application. Figure 6 As shown, the intelligent scheduling method includes the following steps: S601, receiving the target task assigned by the core intelligent agent in the intelligent scheduling system according to the target work plan, the target work plan is generated by the core intelligent agent according to the target interaction information input by the target user, and the target work plan includes at least one target task and target execution information corresponding to at least one target task.

[0070] Specifically, the above-mentioned target agent module may include but is not limited to at least one of the following: a single agent, an agent chain consisting of multiple agents connected in series, and an agent group consisting of multiple agent chains; the above-mentioned target execution information may include but is not limited to at least one of the following: the target task type corresponding to the target task, and the target execution module identifier.

[0071] S602, execute the target task and obtain the corresponding target execution result.

[0072] Optionally, after receiving the target task, the target agent module can directly call the target large model to execute the target task and obtain the corresponding target execution result.

[0073] Optionally, when the target agent module includes multiple target agents, after the target agent module receives the target task, the central agent or any target agent in the target agent module can first disassemble the target task to obtain multiple target subtasks, and then assign each target subtask to the corresponding target agent in the target agent module for execution, and obtain the target sub-execution result corresponding to each target subtask, and finally determine the target execution result corresponding to the target task according to the target sub-execution results corresponding to each of the multiple target subtasks. The above-mentioned target subtask assignment process is similar to the target task assignment process in S202, and will not be repeated here.

[0074] S603, sending the target execution result to the core agent, so that the core agent determines the target interaction result corresponding to the target interaction information according to the target execution result.

[0075] Specifically, after the target intelligent agent module obtains the target execution result, it can directly send the target execution result to the core intelligent agent, or it can only feedback the execution completion status corresponding to the target task to the core intelligent agent, and send the target execution result to the target intelligent agent module that needs to execute the next target task. The embodiment of the present application is not limited to this.

[0076] In the embodiment of the present application, the target agent module in the intelligent scheduling system can be used to assist the core agent in completing the target task corresponding to the target interaction information, thereby improving the intelligent interaction efficiency and intelligence of the intelligent scheduling system.

[0077] In some possible embodiments, Figure 7As shown, after the core agent receives the target interaction information input by the target user, it can generate a corresponding target work plan according to the target interaction information, and call at least one agent module that has reached cooperation or is associated with it to execute the target task in the target work plan to assist it in completing the above target work plan. In the software development scenario, the agent module may include but is not limited to the functions of demand generation, architecture generation, task decomposition, code generation, test case generation, and question and answer. The functions of different agent modules may be the same or different. After the agent module completes the target task, the corresponding target execution result can be output to the core agent. Finally, the core agent can determine the complete target interaction result based on the target execution result of each target task, and output the target interaction result for feedback to the target user.

[0078] It is understandable that the functions of the intelligent agent module may be different in different interaction scenarios. For example, in an educational interaction scenario, the intelligent agent module can have but is not limited to the functions of problem analysis, question generation, answer generation, knowledge point explanation video generation, homework grading, etc.

[0079] Please refer to the following Figure 8 , which is a schematic diagram of the structure of an intelligent scheduling device provided by an exemplary embodiment of the present application. The above-mentioned intelligent scheduling device is applied to the core intelligent body in the intelligent scheduling system, such as Figure 8 As shown, the intelligent scheduling device 800 includes: An acquisition module 810 is used to acquire target interaction information input by a target user; The plan generation module 820 is used to generate a target work plan according to the target interaction information; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: the target task type corresponding to the target task and the target execution module identifier; The calling module 830 is used to call the corresponding target agent module in the above-mentioned intelligent scheduling system to execute the above-mentioned target task according to the target execution information corresponding to the above-mentioned target task, and obtain the corresponding target execution result; the above-mentioned target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; The determination module 840 is used to determine the target interaction result corresponding to the target interaction information according to the target execution result.

[0080] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes target execution orders corresponding to the multiple target tasks. The plan generation module 820 is specifically used for: The target interaction information is analyzed to obtain a total target interaction task; the total target interaction task is decomposed to obtain the multiple target tasks; and the target execution order corresponding to each of the multiple target tasks is determined according to the target task types corresponding to each of the multiple target tasks.

[0081] The calling module 830 is specifically used for: According to the target task types and / or target execution module identifiers corresponding to the above-mentioned multiple target tasks, the corresponding target agent modules in the above-mentioned intelligent scheduling system are called in sequence in the above-mentioned target execution order to execute the above-mentioned target tasks, so as to obtain the target execution results corresponding to the above-mentioned multiple target tasks.

[0082] In a possible implementation, the plan generation module 820 is further used to: According to the target task type corresponding to the above-mentioned target task and the functions and / or status information corresponding to each intelligent agent module in the above-mentioned intelligent scheduling system, the target intelligent agent module that executes the above-mentioned target task is determined; the above-mentioned target execution module identifier is the module identifier corresponding to the above-mentioned target intelligent agent module.

[0083] In a possible implementation, the plan generation module 820 is specifically used to: The target interaction information is analyzed to obtain a target interaction task; according to preset planning rules, a target work plan is determined based on the target type corresponding to the target interaction task; different target types correspond to different target work plans.

[0084] In a possible implementation, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution class and a dependent execution class; the calling module 830 is specifically used to: According to the target task type and / or target execution module identifier corresponding to the first target task, the first target task is assigned to the corresponding first target agent module in the above-mentioned intelligent scheduling system, and the first target execution result output by the above-mentioned first target agent module is obtained; the above-mentioned first target task is a target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks; In the case where there is a second target task among the plurality of target tasks, target input information corresponding to the second target task is generated according to the execution result of the first target on which the second target task depends; the second target task is a target task belonging to the dependent execution class; According to the target task type and / or target execution module identifier corresponding to the above-mentioned second target task, the above-mentioned second target task and the above-mentioned target input information are sent to the corresponding second target intelligent agent module in the above-mentioned intelligent scheduling system, so that the above-mentioned second target intelligent agent module executes the above-mentioned second target task according to the above-mentioned target input information, and obtains the second target execution result output by the above-mentioned second target intelligent agent module.

[0085] In one possible implementation, the intelligent scheduling system includes multiple intelligent agent modules, and the database of the core intelligent agent stores at least one of the following shared information: task execution information corresponding to each of the multiple intelligent agent modules, system information corresponding to the intelligent scheduling system, and task auxiliary information.

[0086] In a possible implementation, the intelligent scheduling device 800 further includes: An information extraction module, used to extract key information and overview information from the shared information when the amount of information corresponding to the shared information is greater than a threshold; The information updating module is used to update the shared information according to the key information and the overview information.

[0087] In a possible implementation, the intelligent scheduling device 800 further includes: The display module is used to display the target execution status and target execution results corresponding to the above-mentioned target work plan according to the current execution status of the above-mentioned target intelligent agent module.

[0088] In a possible implementation, the target agent modules are different or the types corresponding to the target agent modules are different, and the target functions corresponding to the target agent modules are different; The above-mentioned target tasks include at least one of the following: requirement generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks and question and answer tasks.

[0089] Please refer to the following Fig. 9 , which is a schematic diagram of the structure of another intelligent scheduling device provided by an exemplary embodiment of the present application. The above intelligent scheduling device is applied to the target intelligent agent module in the intelligent scheduling system; the above target intelligent agent module includes at least one of the following: a single intelligent agent, an intelligent agent chain composed of multiple intelligent agents in series, and an intelligent agent group composed of multiple such intelligent agent chains; Fig. 9 As shown, the intelligent scheduling device 900 includes: The receiving module 910 is used to receive the target task assigned by the core agent in the intelligent scheduling system according to the target work plan; the target work plan is generated by the core agent according to the target interaction information input by the target user; the target work plan includes at least one of the target task and the target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: the target task type corresponding to the target task and the target execution module identifier; The execution module 920 is used to execute the above target tasks and obtain corresponding target execution results; The sending module 930 is used to send the above-mentioned target execution result to the above-mentioned core intelligent agent, so that the above-mentioned core intelligent agent determines the target interaction result corresponding to the above-mentioned target interaction information according to the above-mentioned target execution result.

[0090] The division of each module in the above-mentioned intelligent scheduling device is only for illustration. In other embodiments, the intelligent scheduling device can be divided into different modules as needed to complete all or part of the functions of the above-mentioned intelligent scheduling device. The implementation of each module in the intelligent scheduling device provided in the embodiments of this specification can be in the form of a computer program. The computer program can be run on a terminal or a server. The program modules constituted by the computer program can be stored in the memory of the terminal or the server. When the computer program is executed by the processor, all or part of the steps of the intelligent scheduling method described in the embodiments of this specification are implemented.

[0091] See next Fig.10 , which is a schematic diagram of the structure of an electronic device provided by an exemplary embodiment of this specification. Fig.10 As shown, the electronic device 1000 may include: at least one processor 1010 , at least one communication bus 1020 , a user interface 1030 , at least one network interface 1040 , and a memory 1050 .

[0092] The communication bus 1020 may be used to realize the connection and communication among the above-mentioned components.

[0093] The user interface 1030 may include a display screen (Display) and a camera (Camera), and the optional user interface 1030 may also include a standard wired interface and a wireless interface.

[0094] The network interface 1040 may optionally include a Bluetooth module, a Near Field Communication (NFC) module, a Wireless Fidelity (Wi-Fi) module, and the like.

[0095] Among them, the processor 1010 may include one or more processing cores. The processor 1010 uses various interfaces and lines to connect various parts of the entire electronic device 1000, and executes various functions and processes data of the routing electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1050, and calling data stored in the memory 1050. Optionally, the processor 1010 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1010 can integrate one or a combination of a processor (Central Processing Unit, CPU), a graphics processor (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1010, but implemented separately through a chip.

[0096] Among them, the memory 1050 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory, ROM). Optionally, the memory 1050 includes a non-transitory computer-readable medium. The memory 1050 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1050 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as an acquisition function, a work plan generation function, an intelligent scheduling function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1050 may also be optionally at least one storage device located away from the aforementioned processor 1010. As Fig.10 As shown, the memory 1050 as a computer storage medium may include an operating system, a network communication module, a user interface module, and program instructions.

[0097] In some possible embodiments, the electronic device 1000 may be the aforementioned Figure 8In the intelligent scheduling device 800 shown, the processor 1010 can be used to call the program instructions stored in the memory 1050, and specifically perform the following operations: obtain target interaction information input by the target user; generate a target work plan based on the above target interaction information; the above target work plan includes at least one target task and target execution information corresponding to the above at least one target task; the above target execution information includes at least one of the following: the target task type and the target execution module identifier corresponding to the above target task; according to the target execution information corresponding to the above target task, call the corresponding target intelligent agent module in the above intelligent scheduling system to execute the above target task and obtain the corresponding target execution result; the above target intelligent agent module includes at least one of the following: a single intelligent agent, an intelligent agent chain composed of multiple intelligent agents in series, and an intelligent agent group composed of multiple such intelligent agent chains; determine the target interaction result corresponding to the above target interaction information based on the above target execution result.

[0098] In some possible embodiments, the target work plan includes multiple target tasks, and the target execution information also includes the target execution order corresponding to each of the multiple target tasks; When the processor 1010 executes the above-mentioned generating a target work plan according to the above-mentioned target interaction information, it is specifically used to execute: The target interaction information is analyzed to obtain a total target interaction task; the total target interaction task is decomposed to obtain the multiple target tasks; and the target execution order corresponding to each of the multiple target tasks is determined according to the target task types corresponding to each of the multiple target tasks.

[0099] The processor 1010 executes the target execution information corresponding to the target task, calls the corresponding target agent module in the intelligent scheduling system to execute the target task, and obtains the corresponding target execution result, which is specifically used to execute: According to the target task types and / or target execution module identifiers corresponding to the above-mentioned multiple target tasks, the corresponding target agent modules in the above-mentioned intelligent scheduling system are called in sequence in the above-mentioned target execution order to execute the above-mentioned target tasks, so as to obtain the target execution results corresponding to the above-mentioned multiple target tasks.

[0100] In some possible embodiments, after the processor 1010 performs the above-mentioned disassembly of the above-mentioned target interaction overall task to obtain the above-mentioned multiple target tasks, it is also used to execute: According to the target task type corresponding to the above-mentioned target task and the functions and / or status information corresponding to each intelligent agent module in the above-mentioned intelligent scheduling system, the target intelligent agent module that executes the above-mentioned target task is determined; the above-mentioned target execution module identifier is the module identifier corresponding to the above-mentioned target intelligent agent module.

[0101] In some possible embodiments, when the processor 1010 executes the above-mentioned generation of the target work plan according to the above-mentioned target interaction information, it is specifically used to execute: The target interaction information is analyzed to obtain a target interaction task; according to preset planning rules, a target work plan is determined based on the target type corresponding to the target interaction task; different target types correspond to different target work plans.

[0102] In some possible embodiments, the target work plan includes multiple target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution type and a dependent execution type; The processor 1010 executes the target execution information corresponding to the target task, calls the corresponding target agent module in the intelligent scheduling system to execute the target task, and obtains the corresponding target execution result, which is specifically used to execute: According to the target task type and / or target execution module identifier corresponding to the first target task, the above-mentioned first target task is assigned to the corresponding first target intelligent agent module in the above-mentioned intelligent scheduling system, and the first target execution result output by the above-mentioned first target intelligent agent module is obtained; the above-mentioned first target task is a target task belonging to the above-mentioned independent execution class among the above-mentioned multiple target tasks.

[0103] In the case that there is a second target task among the above-mentioned multiple target tasks, the target input information corresponding to the above-mentioned second target task is generated according to the execution result of the first target on which the above-mentioned second target task depends; the above-mentioned second target task is a target task belonging to the above-mentioned dependent execution class.

[0104] According to the target task type and / or target execution module identifier corresponding to the above-mentioned second target task, the above-mentioned second target task and the above-mentioned target input information are sent to the corresponding second target intelligent agent module in the above-mentioned intelligent scheduling system, so that the above-mentioned second target intelligent agent module executes the above-mentioned second target task according to the above-mentioned target input information, and obtains the second target execution result output by the above-mentioned second target intelligent agent module.

[0105] In some possible embodiments, the intelligent scheduling system includes multiple intelligent agent modules, and the database of the core intelligent agent stores at least one of the following shared information: task execution information corresponding to each of the multiple intelligent agent modules, system information corresponding to the intelligent scheduling system, and task auxiliary information.

[0106] In some possible embodiments, the processor 1010 is further configured to execute: When the amount of information corresponding to the shared information is greater than a threshold, the key information and overview information in the shared information are extracted; and the shared information is updated according to the key information and the overview information.

[0107] In some possible embodiments, the processor 1010 executes the target execution information corresponding to the target task, calls the corresponding target intelligent agent module in the intelligent scheduling system to execute the target task, and is also used to execute: display the target execution status and target execution result corresponding to the target work plan according to the current execution status of the target intelligent agent module.

[0108] In some possible embodiments, the target agent modules are different or the types corresponding to the target agent modules are different, and the target functions corresponding to the target agent modules are different; The above-mentioned target tasks include at least one of the following: requirement generation tasks, architecture generation tasks, task decomposition tasks, code generation tasks, test case generation tasks and question and answer tasks.

[0109] In some possible embodiments, the electronic device 1000 may be the aforementioned Fig. 9 In the intelligent scheduling device 900 shown, the processor 1010 can be used to call the program instructions stored in the memory 1050, and specifically perform the following operations: receiving the target task assigned by the core intelligent agent in the above-mentioned intelligent scheduling system according to the target work plan; the above-mentioned target work plan is generated by the above-mentioned core intelligent agent according to the target interaction information input by the target user; the above-mentioned target work plan includes at least one of the above-mentioned target tasks and the target execution information corresponding to the above-mentioned at least one target task; the above-mentioned target execution information includes at least one of the following: the target task type corresponding to the above-mentioned target task, and the target execution module identifier; executing the above-mentioned target task to obtain the corresponding target execution result; sending the above-mentioned target execution result to the above-mentioned core intelligent agent, so that the above-mentioned core intelligent agent determines the target interaction result corresponding to the above-mentioned target interaction information according to the above-mentioned target execution result.

[0110] The embodiment of the present application also provides a computer storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes one or more steps in any of the above methods. If the components of the above intelligent scheduling device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above storage medium.

[0111] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The above computer program product includes one or more computer instructions. When the above computer program instructions are loaded and executed on a computer, the above process or function according to the embodiment of the present application is generated in whole or in part. The above computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The above computer instructions may be stored in a computer-readable storage medium or transmitted through the above computer-readable storage medium. The above computer instructions may be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The above computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The above available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.

[0112] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes. In the absence of conflict, the technical features in this embodiment and the implementation scheme can be combined arbitrarily.

[0113] The above-mentioned embodiments are merely preferred embodiments of the present application and are not intended to limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements made to the technical solutions of the present application by ordinary technicians in this field should fall within the protection scope determined by the claims of the present application.

Claims

1. An intelligent scheduling method, characterized in that: The method is applied to the core agent in the intelligent scheduling system; the method comprises: Obtain target interaction information input by the target user; Generate a target work plan according to the target interaction information; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: a target task type corresponding to the target task and a target execution module identifier; According to the target execution information corresponding to the target task, the corresponding target agent module in the intelligent scheduling system is called to execute the target task to obtain the corresponding target execution result; the target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; A target interaction result corresponding to the target interaction information is determined according to the target execution result.

2. The method according to claim 1, characterized in that The target work plan includes a plurality of target tasks, and the target execution information also includes a target execution order corresponding to each of the plurality of target tasks; The generating a target work plan according to the target interaction information includes: Analyzing the target interaction information to obtain a total target interaction task; Decomposing the target interaction total task to obtain the multiple target tasks; Determining a target execution order corresponding to each of the multiple target tasks according to the target task types corresponding to each of the multiple target tasks; The step of calling the corresponding target agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task, and obtaining the corresponding target execution result includes: According to the target task types and / or target execution module identifiers corresponding to each of the multiple target tasks, the corresponding target agent modules in the intelligent scheduling system are called in sequence according to the target execution order to execute the target tasks, so as to obtain the target execution results corresponding to each of the multiple target tasks.

3. The method according to claim 2, characterized in that After the target interaction total task is decomposed to obtain the multiple target tasks, the method further includes: According to the target task type corresponding to the target task and the functions and / or status information corresponding to each intelligent agent module in the intelligent scheduling system, the target intelligent agent module that executes the target task is determined; the target execution module identifier is the module identifier corresponding to the target intelligent agent module.

4. The method according to claim 1, characterized in that: The generating a target work plan according to the target interaction information includes: Analyzing the target interaction information to obtain a total target interaction task; According to the preset planning rules, the target work plan is determined according to the target type corresponding to the target interaction overall task; the target type is different, and the target work plan corresponding to the target interaction overall task is different.

5. The method according to claim 1, characterized in that The target work plan includes a plurality of target tasks, and the target execution information also includes a target execution type corresponding to the target task; the target execution type includes at least one of the following: an independent execution type and a dependent execution type; The step of calling the corresponding target agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task, and obtaining the corresponding target execution result includes: According to the target task type and / or target execution module identifier corresponding to the first target task, the first target task is assigned to the corresponding first target agent module in the intelligent scheduling system, and a first target execution result output by the first target agent module is obtained; the first target task is a target task belonging to the independent execution class among the multiple target tasks; In the case where there is a second target task among the multiple target tasks, generating target input information corresponding to the second target task according to the first target execution result on which the second target task depends; the second target task is a target task belonging to the dependent execution class; According to the target task type and / or target execution module identifier corresponding to the second target task, the second target task and the target input information are sent to the corresponding second target intelligent agent module in the intelligent scheduling system, so that the second target intelligent agent module executes the second target task according to the target input information and obtains the second target execution result output by the second target intelligent agent module.

6. The method according to any one of claims 1 to 5, characterized in that: The target agent modules are different or the types corresponding to the target agent modules are different, the target functions corresponding to the target agent modules are different, and the target tasks include at least one of the following: a demand generation task, an architecture generation task, a task decomposition task, a code generation task, a test case generation task, and a question-answering task; and / or The intelligent scheduling system includes multiple intelligent agent modules, and the database of the core intelligent agent stores at least one of the following shared information: task execution information corresponding to each of the multiple intelligent agent modules, system information corresponding to the intelligent scheduling system, and task auxiliary information.

7. An intelligent scheduling system, characterized in that: The intelligent scheduling system includes a core agent and multiple agent modules; the agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents in series, and an agent group composed of multiple such agent chains; The core agent is used to obtain target interaction information input by a target user, and generate a target work plan according to the target interaction information, wherein the target work plan includes at least one target task and target execution information corresponding to the at least one target task; and, according to the target work plan, call a corresponding target agent module in the intelligent scheduling system to execute the target task and obtain a corresponding target execution result; and, according to the target execution result, determine a target interaction result corresponding to the target interaction information; The agent module is used to execute the tasks assigned by the core agent.

8. An intelligent scheduling device, characterized in that: The intelligent scheduling device is applied to the core intelligent body in the intelligent scheduling system; The intelligent scheduling device comprises: An acquisition module, used to acquire target interaction information input by a target user; A plan generation module, configured to generate a target work plan according to the target interaction information; the target work plan includes at least one target task and target execution information corresponding to the at least one target task; the target execution information includes at least one of the following: a target task type corresponding to the target task and a target execution module identifier; A calling module is used to call the corresponding target agent module in the intelligent scheduling system to execute the target task according to the target execution information corresponding to the target task, and obtain the corresponding target execution result; the target agent module includes at least one of the following: a single agent, an agent chain composed of multiple agents connected in series, and an agent group composed of multiple such agent chains; A determination module is used to determine a target interaction result corresponding to the target interaction information according to the target execution result.

9. An electronic device, characterized in that: include: processor and memory; wherein, The processor is connected to the memory; the memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method steps as described in any one of claims 1-6.

10. A computer storage medium, characterized in that: The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1-6.

Images