CN-122027613-A - Intelligent interaction method and computing device

CN122027613ACN 122027613 ACN122027613 ACN 122027613ACN-122027613-A

Abstract

The embodiment of the application provides an agent interaction method and computing equipment. The method comprises the steps of obtaining an agent event generated by an agent, conducting semantic mapping processing on the agent event to obtain a standard agent user interaction protocol AG-UI event stream, conducting optimization processing on the standard AG-UI event stream according to obtained transmission strategy parameters to obtain a transmission data stream and transmission control information, conducting routing processing on the transmission data stream according to context information corresponding to the obtained agent event to determine a target AG-UI client, and sending the transmission data stream to the target AG-UI client according to the transmission control information so that the target AG-UI client can render a front end interface based on the transmission data stream. The interaction efficiency is improved.

Inventors

Tian Dengkui
KONG WEIJIE

Assignees

河南秦尉数字技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260104

Claims (10)

1. An agent interaction method, comprising: Acquiring an agent event generated by an agent; carrying out semantic mapping processing on the agent event to obtain a standard agent user interaction protocol AG-UI event stream; according to the acquired transmission strategy parameters, optimizing the standard AG-UI event stream to obtain a transmission data stream and transmission control information; According to the acquired context information corresponding to the agent event, carrying out routing processing on the transmission data stream, and determining a target AG-UI client; And sending the transmission data stream to the target AG-UI client according to the transmission control information so that the target AG-UI client can render a front-end interface based on the transmission data stream.
2. The method according to claim 1, wherein the optimizing the standard AG-UI event stream according to the acquired transmission policy parameters to obtain a transmission data stream and transmission control information includes: performing snapshot increment processing on the standard AG-UI event stream to obtain the transmission data stream; and determining the transmission control information according to the transmission strategy parameters.
3. The method of claim 2, wherein said determining said transmission control information based on said transmission policy parameters comprises: and determining a transmission protocol in the transmission control information and a transmission priority in the transmission control information according to the network quality parameter in the transmission strategy parameters.
4. The method of claim 3, wherein said determining a transmission protocol in said transmission control information and a transmission priority in said transmission control information based on a network quality parameter of said transmission policy parameters comprises: Determining a network state according to the delay time length in the network quality parameter and/or the packet loss rate in the network quality parameter; inquiring the corresponding relation between the network state and the transmission protocol according to the network state, and determining the transmission protocol; And determining the transmission priority according to the delay time length in the network quality parameter.
5. The method of claim 2, wherein said determining said transmission control information based on said transmission policy parameters comprises: and determining transmission queue information in the transmission control information according to the service risk level in the transmission strategy parameters, wherein the transmission queue information comprises a queue identifier of a transmission queue and a bandwidth adjustment amplitude of the transmission queue.
6. The method of claim 2, wherein said determining said transmission control information based on said transmission policy parameters comprises: Determining event priority according to the event type in the transmission strategy parameters; And determining a transmission mode in the transmission control information according to the event priority, wherein the transmission mode comprises any one of an instant transmission mode, a semantic slice transmission mode and an aggregation transmission mode.
7. The method according to any one of claims 1-6, wherein said performing semantic mapping on said agent event to obtain a standard agent user interaction protocol AG-UI event stream comprises: Based on a semantic hash index library, carrying out semantic mapping processing on the agent event to obtain a semantic mapping result; And based on a state change event generator, converting the semantic mapping result to obtain the standard AG-UI event stream.
8. The method according to any one of claims 1-7, wherein the routing the transport data stream according to the acquired context information corresponding to the agent event, and determining the target AG-UI client includes: Acquiring the data type in the context information and service logic information in the context information; Determining at least one AG-UI client to be selected from a plurality of initial AG-UI clients according to the data type and the business logic information; Acquiring position information of a processor in the context information; And determining the target AG-UI client from the at least one AG-UI client to be selected according to the position information of the processor, the position information of the processor corresponding to the at least one AG-UI client to be selected and the topological relation of the processor.
9. The method according to any one of claims 1-8, wherein the sending the transport data stream to the target AG-UI client according to the transmission control information comprises: And sending the transmission data stream to the target AG-UI client through a connection channel corresponding to the target AG-UI client according to the transmission control information, wherein the connection channel is pre-established.
10. A computing device, comprising: A processor, and a memory communicatively coupled to the processor; The memory is used for storing computer execution instructions; The processor is configured to execute computer-executable instructions stored in the memory to implement the method of any one of claims 1-9.

Description

Intelligent interaction method and computing device Technical Field The application relates to the technical field of computing equipment, in particular to an agent interaction method and computing equipment. Background In supply chain planning management, enterprises face a complex and diverse supply chain environment. In the traditional supply chain management process, manual inspection of mass data is required, and when the inspection finds risks, a risk conduction path and a coping strategy are manually analyzed. In the related art, a multi-agent may be used to manage a supply chain plan. However, due to the isomerism of the agent framework, the redundancy of the transmission mechanism, and the isolation of the multi-modal data management, the interaction of the agent with the AG-UI client (front-end application) is inefficient, which is a problem in the related art. Disclosure of Invention The embodiment of the application provides an agent interaction method and computing equipment, which can improve interaction efficiency. In a first aspect, an embodiment of the present application provides an agent interaction method, including: Acquiring an agent event generated by an agent; Semantic mapping processing is carried out on the agent event to obtain a standard agent user interaction protocol AG-UI event stream; according to the acquired transmission strategy parameters, optimizing the standard AG-UI event stream to obtain a transmission data stream and transmission control information; according to the acquired context information corresponding to the agent event, carrying out routing processing on the transmission data stream, and determining a target AG-UI client; and sending the transmission data stream to the target AG-UI client according to the transmission control information so that the target AG-UI client can render the front-end interface based on the transmission data stream. In the scheme, the computing equipment can acquire the agent event generated by the agent, and perform semantic mapping processing on the agent event to obtain a standard AG-UI event stream. The computing device can perform optimization processing on the standard AG-UI event stream according to the acquired transmission strategy parameters to obtain a transmission data stream and transmission control information. The computing device may route the transport data stream according to the acquired context information corresponding to the agent event, determine a target AG-UI client, and send the transport data stream to the target AG-UI client according to the transport control information, so that the target AG-UI client renders the front-end interface based on the transport data stream. Through the mode of converting the intelligent agent event into the standard AG-UI event stream, a large number of custom front ends (AG-UI clients) are not needed when enterprises use a plurality of intelligent agent frames in a mixed mode, so that development cost is reduced, and development efficiency is improved. By means of the method for optimizing the standard AG-UI event stream according to the transmission strategy parameters, consumption of transmission bandwidth can be reduced, interaction delay is reduced, and interaction experience of users is improved. By means of the method for routing the transmission data stream according to the context information corresponding to the agent event, data coordination and state synchronization of the cross-equipment are achieved, and interaction delay is further reduced. In conclusion, through the mode, interaction efficiency can be improved. In one implementation, according to the acquired transmission policy parameters, the optimization processing is performed on the standard AG-UI event stream to obtain a transmission data stream and transmission control information, including: Performing snapshot increment processing on the standard AG-UI event stream to obtain a transmission data stream; And determining transmission control information according to the transmission strategy parameters. In this scheme, the computing device may perform snapshot incremental processing on the standard AG-UI event stream to obtain a transport data stream. The computing device may determine the transmission control information based on the transmission policy parameters. By performing snapshot increment processing on the standard AG-UI event stream to obtain the transmission data stream, the data transmission quantity can be reduced, the transmission bandwidth is further saved, and the transmission efficiency is improved. In addition, by determining the transmission control information according to the transmission policy parameters, the transmission delay is reduced, that is, the interaction delay is reduced. In one implementation, determining transmission control information according to transmission policy parameters includes: And determining a transmission protocol in the transmission control information