Man-machine interaction method of charging station management system, electronic equipment and storage medium

Abstract

The invention relates to the field of intelligent interaction, and discloses a man-machine interaction method of a charging station management system, electronic equipment and a storage medium. The method comprises the following steps: a user intelligent agent obtains a user demand input by a user through a user terminal, and converts the user demand into structured data carrying a context identifier; the coordination agent carries out task allocation through an agent group configuration strategy based on the structured data, and functional agents in an agent group are scheduled to execute the allocated tasks; the functional agent executes the task allocated by the coordination agent and returns a completed task result to the coordination agent; and the coordination agent performs consistency verification on the task result returned by the function agent, and performs man-machine interaction with the user through the user agent based on the consistency verification result. Therefore, by constructing multi-agent collaborative man-machine interaction, the charging station management system has higher flexibility, expandability and higher intelligent level.

Description

Technical field

[0001] The present invention relates to the field of intelligent interaction technology, and in particular to a human-computer interaction method, electronic equipment, and storage medium for a charging station management system. [Background Technology]

[0002] With the explosive growth of the electric vehicle industry, electric vehicle charging services have become a core link in the new energy transportation ecosystem.

[0003] As the number of electric vehicles continues to rise, improving charging station operational efficiency, reducing maintenance costs, and increasing user retention have become key issues. Users encounter a wide variety of issues during charging, ranging from charging equipment failures to billing questions, charging station inquiries, and charging policy consultations. The diversity and professionalism of these needs pose significant challenges to the Charging Station Management System (CSMS).

[0004] Currently, existing charging station management systems use AI (Artificial Intelligence) to assist charging station operations and improve operational efficiency. However, because existing AI generally adopts a single-agent architecture and the interaction mode is limited to a "one-question-one-answer" model, charging station management systems have many shortcomings when facing a growing number of users, such as:

[0005] 1. Lack of flexibility: It is impossible to dynamically call multiple professional agents to collaborate according to scenario requirements, making it difficult to flexibly respond to complex and changing operational scenarios;

[0006] 2. Weak information integration capabilities: Difficulty handling complex problems, such as those requiring multiple rounds of reasoning or multi-source data fusion;

[0007] 3. Passive interaction: Relies on users to actively provide complete information and cannot actively guide the conversation.

[0008] Therefore, when facing an increasingly large user base, the existing charging station management system with a single agent architecture has unstable service quality and is difficult to achieve all-weather support. It also increases the operating costs of the charging station and reduces the overall economic benefits of the charging station. [Summary of the invention]

[0009] The embodiments of the present invention aim to provide a human-computer interaction method, electronic device, and storage medium for a charging station management system, aiming to solve the problems of insufficient flexibility and weak information integration capabilities of the existing charging station management system with a single agent architecture.

[0010] To solve the above technical problems, a first embodiment of the present invention provides a human-computer interaction method for a charging station management system, comprising:

[0011] The user agent obtains the user demand input by the user through the user terminal, converts the user demand into structured data carrying a context identifier, and sends the structured data to the coordination agent;

[0012] The coordinating agent allocates tasks based on the structured data through an agent group configuration strategy, and schedules the functional agents in the agent group to perform the allocated tasks;

[0013] The functional agent performs the task assigned by the coordination agent and returns the completed task result to the coordination agent;

[0014] The coordination agent performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent based on the consistency verification result.

[0015] Optionally, the context identifier includes a unique session identifier, a message sequence number, a semantic summary, and an upstream task ID on which it depends; the structured data includes a message ID, a message type, and a message body, and the context identifier is stored in the message body.

[0016] Optionally, the coordination agent allocates tasks based on the structured data through an agent group configuration strategy, and schedules functional agents in the agent group to perform the allocated tasks, including:

[0017] The coordination agent parses the structured data and extracts context identifiers from the structured data;

[0018] The coordination agent determines the user demand input by the user according to the context identifier through a preset AI model;

[0019] According to the determined user needs, tasks are allocated through the agent group configuration strategy, and the functional agents in the agent group are scheduled to perform the allocated tasks.

[0020] Optionally, the coordination agent determines the user demand input by the user according to the context identifier through a preset AI model, including:

[0021] The coordination agent parses the semantic summary in the context identifier through a preset AI model, identifies the semantic summary through the preset AI model, and determines the user requirements input by the user.

[0022] Optionally, the agent group configuration strategy includes: a selection mode and a polling mode;

[0023] When the selection mode is used, the coordination agent selects a single optimal functional agent to perform the task according to the Q-Learning dynamic scheduling allocation algorithm;

[0024] When the polling mode is used, the coordinating agent broadcasts the task to all functional agents in the agent group. Each functional agent decides whether to participate through capability self-inspection. The coordinating agent calls each functional agent participating in the task to participate in the execution of the task in sequence.

[0025] Optionally, performing task allocation according to the determined user needs through an agent group configuration strategy and scheduling functional agents in the agent group to perform the allocated tasks includes:

[0026] If the user demand determined by the coordinating agent is related to a functional agent in the agent group, the selection mode in the agent group configuration strategy is called to call the corresponding functional agent in the agent group, assign the task to the functional agent to perform the task, and return the task result to the coordinating agent;

[0027] If the user needs determined by the coordinating agent need to be completed by a combination of multiple agents, the user needs are split into multiple subtasks through the preset AI model, and the polling mode in the agent group configuration strategy is called, and multiple corresponding functional agents in the agent group are called to participate in sequence, and each subtask is assigned to the corresponding functional agent in the agent group to perform the task. Each participating functional agent returns its own task results to the coordinating agent for integration.

[0028] Optionally, before the coordination agent assigns tasks, the human-computer interaction method of the charging station management system further includes:

[0029] The coordinating agent queries the context memory library to see whether there is a historical state related to this task. If so, the task is updated based on the historical state returned by the context memory library, and then the updated task is executed through the agent group configuration strategy; if not, the task is executed through the agent group configuration strategy.

[0030] Optionally, the coordination agent performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent based on the consistency verification result, including:

[0031] If the consistency check passes, it means that the task is completed normally, and the coordination agent returns the integrated task results to the user agent;

[0032] If the consistency check fails, it means that the task is not completed, triggering an active inquiry mechanism or a task reallocation mechanism. The coordination agent reallocates the task and returns the task result to the user agent.

[0033] Accordingly, an embodiment of the second aspect of the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and running on the processor. When the computer program is executed by the processor, the human-computer interaction method of the charging station management system described in the embodiment of the first aspect of the present invention is implemented.

[0034] Accordingly, an embodiment of the third aspect of the present invention provides a storage medium, on which a program of a human-computer interaction method for a charging station management system is stored. When the program of the human-computer interaction method for a charging station management system is executed by a processor, the human-computer interaction method for a charging station management system described in the embodiment of the first aspect of the present invention is implemented.

[0035] Compared with the prior art, an embodiment of the present invention provides a human-computer interaction method, electronic device and storage medium for a charging station management system. The human-computer interaction method for a charging station management system includes: a user agent obtains user requirements input by a user through a user terminal, and converts the user requirements into structured data carrying a context identifier, and sends it to a coordination agent; the coordination agent performs task allocation based on the structured data through an agent group configuration strategy, and schedules functional agents in the agent group to perform the assigned tasks; the functional agent performs the task assigned by the coordination agent, and returns the completed task result to the coordination agent; the coordination agent performs consistency verification on the task result returned by the functional agent, and performs human-computer interaction with the user through the user agent based on the consistency verification result. By building a multi-agent collaborative human-computer interaction model, users can be incorporated into human-computer interaction as agents, enabling dynamic and seamless collaboration between humans and AI. This breaks the single question-and-answer model and supports hybrid interaction logic such as one question with multiple answers, multiple questions with one answer, and proactive inquiry. Furthermore, multi-agent collaboration can decouple complex user needs into different functional agents and enable them to collaborate with each other, making the charging station management system with multi-agent collaborative agents more flexible, scalable, and intelligent. This improves information integration capabilities, reduces the operational management burden of charging stations, enables intelligent decision-making and policy execution for operational management, enhances intelligent decision-making capabilities for operational management, and improves the overall revenue and profitability of charging stations. This addresses the current problems of insufficient flexibility and weak information integration capabilities in charging station management systems with a single-agent architecture.

Brief Description of the Drawings

[0036] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings. These exemplifications do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0037] Figure 1 This is a structural diagram of a human-computer interaction system of a charging station management system provided by the present invention;

[0038] Figure 2 This is a flow chart of a human-computer interaction method for a charging station management system provided by the present invention;

[0039] Figure 3 This is a detailed flowchart of a human-computer interaction method for a charging station management system provided by the present invention;

[0040] Figure 4 This is a schematic diagram of the relationship between a context identifier and a context memory library in a human-computer interaction method for a charging station management system provided by the present invention;

[0041] Figure 5 It is a structural schematic diagram of an electronic device provided by the present invention. [Specific implementation method]

[0042] In order to facilitate the understanding of the present invention, the present invention will be described in more detail below with reference to the accompanying drawings and specific embodiments. It should be noted that when an element is described as being "fixed to" another element, it can be directly on the other element, or there can be one or more centered elements therebetween. When an element is described as being "electrically connected" to another element, it can be directly connected to the other element, or there can be one or more centered elements therebetween. The orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", "bottom" etc. used in this specification is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0043] Unless otherwise defined, all technical and scientific terms used in this specification have the same meanings as those commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification are intended only to describe specific embodiments and are not intended to limit the invention. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0044] In addition, the technical features involved in different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0045] In the present invention, Figure 1 As shown, a human-computer interaction system 100 of a charging station management system based on multi-agent collaboration is provided, including: an electronic device 900, in which a human-computer interaction method of a charging station management system provided by the present invention is integrated.

[0046] The electronic device 900 may be any device equipped with a processor and having processing capabilities, such as a smart phone, a tablet computer, a PDA, a laptop computer, a server, a workstation, or other electronic device equipped with a processor.

[0047] In this invention, Agent (or intelligent agent) is a module with autonomous decision-making capabilities, which can plan tasks, call tools (such as APIs, databases), and iterate and optimize results. Specifically, it includes:

[0048] Autonomous decision-making ability: Ability to make decisions and take actions independently under the constraints of specified goals or rules based on perceived environmental information and existing knowledge base.

[0049] Communication and collaboration: Multiple agents can communicate with each other through message queues, RPC (remote procedure calls), HTTP / HTTPS-based APIs, etc., to obtain task-related context information and collaboratively complete complex tasks.

[0050] Adaptive learning: Agents can continuously update strategies based on historical data and real-time feedback through algorithms such as machine learning and reinforcement learning, thereby improving decision-making quality.

[0051] In one embodiment, Figure 2 and Figure 3 As shown, the present invention provides a human-computer interaction method for a charging station management system, based on multi-agent collaboration. The method is applied to the charging station management system, including:

[0052] S1. The user agent (user agent) 20 obtains the user demand input by the user through the user terminal 10, converts the user demand into structured data carrying a context identifier, and sends it to the coordination agent (coordination agent) 30;

[0053] S2. The coordination agent 30 allocates tasks based on the structured data through the agent group 40 configuration strategy, and schedules the functional agents in the agent group 40 to perform the assigned tasks;

[0054] S3, the functional agent performs the task assigned by the coordination agent and returns the completed task results to the coordination agent 30;

[0055] S4. The coordination agent 30 performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent 20 based on the consistency verification result.

[0056] In this embodiment, a human-computer interaction method for a charging station management system based on multi-agent collaboration is provided, including: the user agent obtains the user needs input by the user through the user terminal, and converts the user needs into structured data carrying context identifiers, and sends it to the coordination agent; the coordination agent allocates tasks based on the structured data through the agent group configuration strategy, and schedules the functional agents in the agent group to perform the assigned tasks; the functional agents perform the tasks assigned by the coordination agent, and return the completed task results to the coordination agent; the coordination agent performs consistency verification on the task results returned by the functional agents, and performs human-computer interaction with the user through the user agent based on the consistency verification results. By building a multi-agent collaborative human-computer interaction model, users can be incorporated into human-computer interaction as agents, enabling dynamic and seamless collaboration between humans and AI. This breaks the single question-and-answer model and supports hybrid interaction logic such as one question with multiple answers, multiple questions with one answer, and proactive inquiry. Furthermore, multi-agent collaboration can decouple complex user needs into different functional agents and enable them to collaborate with each other, making the charging station management system with multi-agent collaborative agents more flexible, scalable, and intelligent. This improves information integration capabilities, reduces the operational management burden of charging stations, enables intelligent decision-making and policy execution for operational management, enhances intelligent decision-making capabilities for operational management, and improves the overall revenue and profitability of charging stations. This addresses the current problems of insufficient flexibility and weak information integration capabilities in charging station management systems with a single-agent architecture.

[0057] In one embodiment, Figure 2 and Figure 3 As shown, in step S1, the user agent 20 obtains the user demand input by the user through the user terminal 10, converts the user demand into structured data carrying a context identifier, and sends it to the coordination agent 30, specifically including:

[0058] S11 , the user agent 20 obtains user requirements input by the user through the user terminal 10 .

[0059] The user inputs user needs through the user terminal 10, and the user needs include charging needs or question consultation needs. The user needs are input in a text, voice and / or image manner.

[0060] The user may input user requirements including text, voice and / or image through the user terminal 10 , and the user agent 20 obtains the user requirements input by the user through the user terminal 10 .

[0061] S12 , the user agent 20 parses the user requirements input by the user, converts the user requirements into structured data carrying a context identifier context_tag, and sends the structured data to the coordination agent 30 .

[0062] The context identifier context_tag includes (as shown in the following table):

[0063] Unique session identifier: session_id, which is used throughout the entire interaction cycle;

[0064] Message sequence number: sequence_num, used for sorting and deduplication;

[0065] Semantic summary: context_hash, semantic summary of the first three messages (SHA-256 compression);

[0066] The dependent upstream task ID: dependency_id, used for multi-level task tracking.

[0067] context identifier (context_tag)

[0068] Field Name Data Type illustrate session_id UUID Unique session identifier throughout the entire interaction cycle sequence_num Integer Message sequence number, used for sorting and deduplication context_hash String Semantic digest of the first 3 messages (SHA-256 compressed) dependency_id UUID The upstream task ID that it depends on, used for multi-level task tracking

[0069] The structured data includes the message ID (message_id), message type (message_type), and message body (message_body):

[0070] Message ID (message_id): used for idempotence control and message tracing;

[0071] Message type (message_type): used as the basis for routing decisions;

[0072] Message body (message_body): used for actual interaction content; the context identifier (context_tag) is stored in the message body.

[0073] Structured data

[0074] Field Name effect Optional Value Examples message_id Idempotence control and message tracing UUIDv4 or timestamp + random number message_type Routing decision basis TASK_REQUEST / QUERY_RESPONSE / ERROR message_body Actual interactive content Text / JSON / binary data

[0075] For example, the structured data table uses JSON as an example to illustrate the storage location of structured data, as shown below:

[0076]

[0077]

[0078] In this embodiment, after receiving a user request input via a user terminal, the user agent further analyzes the request, converting it into structured data carrying a context identifier. A message ID is automatically generated for the request, and the specific content of the request is included in the message body, thereby forming structured data carrying a context identifier. The agent then exchanges context information with other agents using the context identifier. This incorporates the user as an agent into human-computer interaction, enabling dynamic collaboration between humans and AI, and seamless human-computer collaboration.

[0079] In one embodiment, Figure 2 and Figure 3 As shown, in step S2, the coordination agent 30 performs task allocation based on structured data through the agent group (Agent Group) configuration strategy, and schedules the functional agents in the agent group 40 to perform the allocated tasks.

[0080] The coordination agent 30 of the present invention can realize the following functions:

[0081] Task analysis and allocation: Analyze user needs, split them into subtasks that can be undertaken by different functional agents, and allocate the subtasks.

[0082] Task result integration and conflict resolution: After each functional agent completes the processing, it will return the results to the coordination agent 30, which will summarize, compare and check conflicts (i.e., consistency verification, such as resource scheduling conflicts, price setting conflicts, etc.). If necessary, it will actively interact with the user again through an active inquiry mechanism to clarify user needs or confirm the final solution.

[0083] Priority control: When resources are limited or demands conflict, the execution order and resource allocation of different functional agents in the agent group 40 are coordinated through predefined priority rules or dynamic decision algorithms (such as hierarchical analysis method, Markov decision process, etc.).

[0084] Functional agents are agents that are assigned roles or have the ability to interact with the environment to solve specific problems. Agent group 40 includes multiple functional agents, such as a scheduling agent (scheduling agent), a billing agent (billing agent), a user service agent (user service agent), and a revenue analysis agent (revenue analysis agent).

[0085] The dispatching agent is used to perform real-time load assessment based on the usage of charging equipment (e.g., DC charging piles, AC charging piles) at the charging station (e.g., the number of vehicles currently charging, queue status), and the status of the charging equipment (e.g., power limit, maintenance status). Specifically, it includes:

[0086] Real-time load analysis: Perform real-time load assessment based on the usage and status of charging equipment at the charging station.

[0087] Scheduling algorithms and strategies: Various scheduling algorithms are used for scheduling, such as the Shortest Job First (SJF) algorithm based on queuing theory, a heuristic algorithm based on profit maximization, or a multi-objective optimization algorithm with vehicle type / battery level / user membership level as weights.

[0088] Cross-site coordination: If a charging operator owns multiple charging stations in different locations, a cross-site scheduling strategy can be adopted to guide some users to relatively idle charging stations to improve overall efficiency.

[0089] The billing agent is used to automatically generate differentiated pricing plans for different users based on a dynamic pricing model, perform real-time billing and settlement when users charge their batteries, and connect with the payment system to complete automatic deductions or bill push. Specifically, it includes:

[0090] Dynamic pricing model: Automatically generates differentiated pricing plans for different users by comprehensively considering factors such as charging time period, peak and valley electricity prices, real-time load, membership level, and promotion strategies.

[0091] Real-time billing and settlement: After the user starts charging, the background system records the charging duration, power consumption, rate changes and other information in real time to generate real-time charging records; when the user finishes charging, the bill is automatically settled and generated.

[0092] Integration with payment systems: Connect with third-party payment platforms (such as WeChat, Alipay or proprietary payment systems) to complete automatic deductions or bill push.

[0093] User service agent, used for personalized service recommendations, multi-channel delivery and proactive notifications; specifically includes:

[0094] Personalized service recommendation: Based on the user's historical charging data, vehicle information and preferences (such as different price sensitivity at different time periods), the system automatically recommends the best or most suitable charging plan when the user makes a charging request.

[0095] Multi-channel interaction: supports multiple interfaces such as the web, vehicle-mounted systems, and even voice interaction to enhance user experience.

[0096] Active notification: Charging progress or cost information can be proactively notified to users via SMS, email push, etc., and additional paid value-added services such as extended charging and reserved parking spaces can be provided when necessary.

[0097] Profit analysis agent, used for historical data analysis, forecasting and strategy optimization, and risk warning; specifically includes:

[0098] Historical data analysis: Based on historical operating data, analyze key information such as user traffic, charging station utilization rate, and revenue in each time period, and form a visual report for operators' reference.

[0099] Forecasting and strategy optimization: Through machine learning models (such as time series analysis, regression prediction, and revenue optimization based on reinforcement learning), we predict future electricity demand and revenue trends and provide strategy optimization suggestions (for example, suggesting appropriate rate increases or specific promotional activities during certain periods of time).

[0100] Risk warning: When abnormal data is detected (such as abnormal user loss, sudden drop in revenue, high equipment abnormality rate), an operational risk warning will be issued in a timely manner.

[0101] Functional agents also have a pluggable agent design, which allows for the flexible addition or removal of agents based on the needs of different operators. For example, a vehicle identification agent or a coupon marketing agent can be added.

[0102] like Figure 2 and Figure 3 As shown, in step S2, the coordination agent 30 allocates tasks based on the structured data through the agent group configuration strategy, and schedules the functional agents in the agent group 40 to perform the assigned tasks; specifically, the steps include:

[0103] S21. The coordination agent 30 parses the structured data and extracts the context identifiers in the structured data.

[0104] The coordination agent 30 receives the structured data carrying the context identifier context_tag sent by the user agent 20, and the coordination agent 30 parses the structured data and extracts the context identifier context_tag from the structured data.

[0105] S22. The coordination agent 30 determines the user requirements input by the user according to the context identifier through a preset AI model.

[0106] The coordination agent 30 parses the semantic summary context_hash in the context identifier context_tag through a preset AI model, identifies the semantic summary context_hash through the preset AI model, and determines the user requirements input by the user.

[0107] For example, the preset AI model includes the Claude-3-7-sonnet-20250219 AI model and may also include the GPT-4o AI model.

[0108] S23. According to the determined user needs, tasks are assigned through the agent group configuration strategy, and the functional agents in the agent group 40 are scheduled to perform the assigned tasks.

[0109] Specifically, the agent group configuration strategies include: selection mode (Select mode) and polling mode (Round mode); when the selection mode is used, the coordinating agent 30 selects a single optimal functional agent to perform the task according to the Q-Learning dynamic scheduling allocation algorithm; when the polling mode is used, the coordinating agent 30 broadcasts the task to all functional agents in the agent group 40, and each functional agent decides whether to participate through self-inspection of its capabilities. The coordinating agent 30 calls each functional agent participating in the task to participate in the execution of the task in sequence.

[0110] Agent Group Configuration Strategy

[0111]

[0112] In step S23, according to the determined user needs, the task is assigned through the agent group configuration strategy, and the functional agents in the agent group 40 are scheduled to perform the assigned tasks; specifically, the following steps are performed:

[0113] S231. If the user demand determined by the coordinating agent 30 is related to a functional agent in the agent group 40, the coordinating agent 30 calls the selection mode in the agent group configuration strategy, calls the corresponding functional agent in the agent group 40, assigns the task to the functional agent to perform the task, and returns the task result to the coordinating agent 30.

[0114] For example, the agent group 40 includes a scheduling agent, a billing agent, a user service agent, and a revenue analysis agent.

[0115] If the user demand determined by the coordinating agent 30 is scheduling-related, the coordinating agent 30 calls the selection mode in the Agent Group configuration strategy, calls the scheduling agent in the agent group 40, assigns the task to the scheduling agent to execute the task, so that the scheduling agent performs resource strategy optimization and returns the task result to the coordinating agent 30.

[0116] If the user demand determined by the coordination agent 30 is related to billing, the coordination agent 30 calls the selection mode in the Agent Group configuration strategy, calls the billing agent in the agent group 40, assigns the task to the billing agent to perform the task, so that the billing agent performs dynamic rate calculation and returns the task result to the coordination agent 30.

[0117] If the user demand determined by the coordination agent 30 is related to profit, the coordination agent 30 calls the selection mode in the Agent Group configuration strategy, calls the profit analysis agent in the agent group 40, assigns the task to the profit analysis agent to perform the task, so that the profit analysis agent performs charging station profit optimization and returns the task result to the coordination agent 30.

[0118] If the user demand determined by the coordination agent 30 is related to the user response, the coordination agent 30 calls the selection mode in the AgentGroup configuration strategy, calls the user service agent in the agent group 40, assigns the task to the user service agent to perform the task, so that the user service agent performs personalized recommendations and returns the task results to the coordination agent 30.

[0119] At this time, when the coordination agent 30 calls the selection mode in the Agent Group configuration strategy, the tasks are scheduled and allocated in combination with the dynamic scheduling allocation algorithm based on Q-Learning to improve the accuracy of task scheduling and allocation.

[0120] The pseudo code of the Q-Learning dynamic scheduling allocation algorithm is as follows:

[0121]

[0122] S232. If the user needs determined by the coordinating agent 30 need to be completed by a combination of multiple agents, the coordinating agent 30 will split the user needs into multiple subtasks through the preset AI model, and call the polling mode in the Agent Group configuration strategy, calling multiple corresponding functional agents in the agent group 40 to participate in sequence, and assigning each subtask to the corresponding functional agent in the agent group 40 to perform the task. Each participating functional agent returns its own task results to the coordinating agent 30 for integration.

[0123] Furthermore, before the coordination agent 30 assigns tasks, the human-computer interaction method of the charging station management system also includes: the coordination agent 30 will first query the context memory 50 to see whether there is a historical state related to the current task. If so, the current task will be updated based on the historical state returned by the context memory 50, and then the updated task will be executed through the Agent Group configuration strategy; if not, the current task will be executed through the Agent Group configuration strategy.

[0124] Specifically, if Figure 4 , which is a schematic diagram of the relationship between the context identifier context_tag and the context memory 50 .

[0125] The context memory library 50 is used to store historical interaction records and charging data.

[0126] The user agent 20 sends the structured data carrying the context identifier context_tag to the coordination agent 30 .

[0127] The coordination agent 30 parses the structured data, extracts the unique session identifier session_id from the context identifier context_tag, and sends it to the context memory 50 to query the historical status of the unique session identifier session_id stored in the context memory 50. After receiving the historical status of the unique session identifier session_id returned by the context memory 50, the semantic summary context_hash in the context identifier context_tag is updated according to the historical status of the returned unique session identifier session_id, and a new message is generated and returned to the user agent 20, which updates the current context identifier context_tag.

[0128] In this embodiment, a coordinated agent parses structured data, extracts contextual identifiers from the structured data, uses a preset AI model to determine user needs inputted based on the contextual identifiers, allocates tasks based on agent group configuration policies, and schedules functional agents within the agent group to execute the assigned tasks. This creates a multi-agent collaborative human-computer interaction model, incorporating users as agents into human-computer interaction, enabling dynamic collaboration between humans and AI, seamless human-computer collaboration, and breaking the single question-and-answer model to support hybrid interaction logic such as one question, multiple answers, multiple questions, and one answer, as well as proactive inquiries. Furthermore, multi-agent collaboration can decouple complex user needs into different functional agents, enabling them to collaborate with each other. This makes the charging station management system, which incorporates a multi-agent collaborative agent group, more flexible, scalable, and intelligent, improving information integration capabilities, reducing the operational management burden of charging stations, enabling intelligent decision-making and policy execution, enhancing intelligent decision-making capabilities, and improving the overall revenue and profitability of charging stations.

[0129] In one embodiment, Figure 2 and Figure 3 As shown, in step S4, the coordination agent 30 performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent 20 based on the consistency verification result.

[0130] Specifically, the coordination agent 30 receives the task results returned by the functional agents, integrates the task results, verifies the consistency of the integrated task results with the user requirements input by the user, and performs the following processing based on the consistency verification results:

[0131] If the consistency check passes, it means that the task is completed normally, and the coordination agent 30 returns the integrated task results to the user agent 20.

[0132] If the consistency check fails, it means that the task is not completed, triggering the active query mechanism or task reallocation mechanism to further determine the reason why the task was not completed:

[0133] If the reason is missing or contradictory information, the coordination agent 30 generates an inquiry request (for example, "Please supplement the time range") and actively asks the user through the active inquiry mechanism. After the coordination agent 30 obtains the user's feedback information through the user agent 20, it re-enters the processing flow. Among them, the active inquiry mechanism actively asks questions to the user agent 20 based on the context to supplement information or clarify intentions.

[0134] If the reason is that the functional agent cannot complete the task, the coordination agent 30 will reallocate the task to other functional agents through the task reallocation mechanism to continue execution; wherein, the task reallocation mechanism is that the coordination agent 30 will reschedule the functional agents in the agent group 40 to perform the unfinished tasks based on user needs and Agent Group configuration strategies; if the task cannot be completed after being reallocated by the task reallocation mechanism, the coordination agent 30 will return the result of the inability to complete the task to the user agent 20.

[0135] If the reason is that the functional agent has only completed part of the subtasks of the task and the remaining subtasks of the task require the cooperation of other functional agents to complete, the coordinating agent 30 will reallocate the remaining subtasks of the task to other functional agents through the task reallocation mechanism to continue execution.

[0136] In the present invention, the consistency check is a semantic consistency check, in which the coordination agent compares whether the content input by the user through the user agent is consistent with the content of the integrated task result. If they are consistent, the integrated task result user agent is returned to the user; if the two are inconsistent within the preset maximum number of rounds (N times), the coordination agent requires other functional agents to continue to complete the task; if the preset maximum number of rounds (N times) is exceeded, the coordination agent notifies the user through the user agent that the task cannot be completed.

[0137] Furthermore, the coordination agent 30 receives the task results returned by the functional agents and integrates the task results, specifically including:

[0138] The coordination agent 30 integrates the task results returned by the multifunctional agent through a preset AI model, and comprehensively outputs the optimal charging plan to the user agent 20. The output format includes text, charts or multimodal combinations.

[0139] In this embodiment, the coordination agent performs consistency verification on the task results returned by the functional agent, and based on the consistency verification results, human-computer interaction is performed with the user through the user agent, thereby improving the accuracy of questions and answers and enhancing the user experience.

[0140] Based on the same concept, the present invention also provides an electronic device 900, such as Figure 1 and Figure 5 As shown, the electronic device 900 includes: a memory 902, a processor 901, and one or more computer programs stored in the memory 902 and executable on the processor 901. The memory 902 and the processor 901 are coupled together via a bus system 903. When the one or more computer programs are executed by the processor 901, the following steps of a human-computer interaction method for a charging station management system provided in an embodiment of the present invention are implemented:

[0141] S1. The user agent 20 obtains the user demand input by the user through the user terminal 10, converts the user demand into structured data carrying a context identifier, and sends it to the coordination agent 30;

[0142] S2. The coordination agent 30 allocates tasks based on the structured data and configures the agent group strategy, scheduling the functional agents in the agent group 40 to perform the assigned tasks.

[0143] S3, the functional agent performs the task assigned by the coordination agent and returns the completed task results to the coordination agent 30;

[0144] S4. The coordination agent 30 performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent 20 based on the consistency verification result.

[0145] The methods disclosed in the above embodiments of the present invention can be applied to or implemented by processor 901. Processor 901 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be performed by hardware integrated logic circuits or software instructions within processor 901. Processor 901 can be a general-purpose processor, a DSP, or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. Processor 901 can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present invention can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules within the decoding processor. The software modules can be located in a storage medium located in memory 902. Processor 901 reads information from memory 902 and, in conjunction with its hardware, completes the steps of the above method.

[0146] It can be understood that the memory 902 in the embodiment of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device; the volatile memory can be random access memory (RAM), by way of example but not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), etc. Memory), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), Direct Rambus Random Access Memory (DRRAM).The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memories.

[0147] It should be noted that the above-mentioned electronic device embodiment and method embodiment belong to the same concept, and their specific implementation process is detailed in the method embodiment, and the technical features in the method embodiment are applicable to the electronic device embodiment, which will not be repeated here.

[0148] In addition, in an exemplary embodiment, an embodiment of the present invention further provides a computer storage medium, specifically a computer-readable storage medium, for example, including a memory 902 storing a computer program, on which one or more programs of a human-computer interaction method for a charging station management system are stored. When the one or more programs of the human-computer interaction method for a charging station management system are executed by the processor 901, the following steps of the human-computer interaction method for a charging station management system provided in an embodiment of the present invention are implemented:

[0149] S1. The user agent 20 obtains the user demand input by the user through the user terminal 10, converts the user demand into structured data carrying a context identifier, and sends it to the coordination agent 30;

[0150] S2. The coordination agent 30 allocates tasks based on the structured data and configures the agent group strategy, scheduling the functional agents in the agent group 40 to perform the assigned tasks.

[0151] S3, the functional agent performs the task assigned by the coordination agent and returns the completed task results to the coordination agent 30;

[0152] S4. The coordination agent 30 performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent 20 based on the consistency verification result.

[0153] It should be noted that the human-computer interaction method program embodiment of the charging station management system on the above-mentioned computer-readable storage medium and the method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment, and the technical features in the method embodiment are correspondingly applicable in the embodiment of the above-mentioned computer-readable storage medium, which will not be repeated here.

[0154] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Under the concept of the present invention, the technical features in the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations in different aspects of the present invention as described above. For the sake of simplicity, they are not provided in detail. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A human-computer interaction method for a charging station management system, characterized in that: include: The user agent obtains the user demand input by the user through the user terminal, converts the user demand into structured data carrying a context identifier, and sends the structured data to the coordination agent; The coordinating agent allocates tasks based on the structured data through an agent group configuration strategy, and schedules the functional agents in the agent group to perform the allocated tasks; The functional agent performs the task assigned by the coordination agent and returns the completed task result to the coordination agent; The coordination agent performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent based on the consistency verification result.

2. The human-computer interaction method for a charging station management system according to claim 1, characterized in that: The context identifier includes a unique session identifier, a message sequence number, a semantic summary, and an upstream task ID on which it depends; the structured data includes a message ID, a message type, and a message body, and the context identifier is stored in the message body.

3. The human-computer interaction method for the charging station management system according to claim 2, characterized in that: The coordination agent allocates tasks based on the structured data through an agent group configuration strategy and schedules functional agents in the agent group to perform the allocated tasks, including: The coordination agent parses the structured data and extracts context identifiers from the structured data; The coordination agent determines the user demand input by the user according to the context identifier through a preset AI model; According to the determined user needs, tasks are allocated through the agent group configuration strategy, and the functional agents in the agent group are scheduled to perform the allocated tasks.

4. The human-computer interaction method for a charging station management system according to claim 3, characterized in that: The coordination agent determines the user demand input by the user according to the context identifier through a preset AI model, including: The coordination agent parses the semantic summary in the context identifier through a preset AI model, identifies the semantic summary through the preset AI model, and determines the user requirements input by the user.

5. The human-computer interaction method for the charging station management system according to claim 3, characterized in that: The agent group configuration strategy includes: selection mode and polling mode; When the selection mode is used, the coordination agent selects a single optimal functional agent to perform the task according to the Q-Learning dynamic scheduling allocation algorithm; When the polling mode is used, the coordinating agent broadcasts the task to all functional agents in the agent group. Each functional agent decides whether to participate through capability self-test. The coordinating agent calls each functional agent participating in the task to participate in the execution of the task in sequence.

6. The human-computer interaction method for the charging station management system according to claim 5, characterized in that: The step of allocating tasks according to the determined user needs through an agent group configuration strategy and scheduling functional agents in the agent group to perform the assigned tasks includes: If the user demand determined by the coordinating agent is related to a functional agent in the agent group, the selection mode in the agent group configuration strategy is called to call the corresponding functional agent in the agent group, assign the task to the functional agent to perform the task, and return the task result to the coordinating agent; If the user needs determined by the coordinating agent need to be completed by a combination of multiple agents, the user needs are split into multiple subtasks through the preset AI model, and the polling mode in the agent group configuration strategy is called, and multiple corresponding functional agents in the agent group are called to participate in sequence, and each subtask is assigned to the corresponding functional agent in the agent group to perform the task. Each participating functional agent returns its own task results to the coordinating agent for integration.

7. The human-computer interaction method for a charging station management system according to claim 1, characterized in that: Before the coordination agent assigns tasks, the human-computer interaction method of the charging station management system further includes: The coordinating agent queries the context memory library to see whether there is a historical state related to this task. If so, the task is updated based on the historical state returned by the context memory library, and then the updated task is executed through the agent group configuration strategy; if not, the task is executed through the agent group configuration strategy.

8. The human-computer interaction method for a charging station management system according to claim 1, characterized in that: The coordination agent performs consistency verification on the task results returned by the functional agent, and performs human-computer interaction with the user through the user agent based on the consistency verification result, including: If the consistency check passes, it means that the task is completed normally, and the coordination agent returns the integrated task results to the user agent; If the consistency check fails, it means that the task is not completed, triggering an active inquiry mechanism or a task reallocation mechanism. The coordination agent reallocates the task and returns the task result to the user agent.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the computer program is executed by the processor, the human-computer interaction method of the charging station management system according to any one of claims 1 to 8 is implemented.

10. A storage medium, characterized in that: The storage medium stores a program of a human-computer interaction method for a charging station management system, and when the program of the human-computer interaction method for a charging station management system is executed by a processor, the human-computer interaction method for a charging station management system according to any one of claims 1 to 8 is implemented.

Images