Multi-modal man-machine interaction platform system and method based on central coordination layer

By using a multimodal human-computer interaction platform system based on a central coordination layer, multimodal inputs are converted into structured semantic event objects, personalized responses are generated, and provided through standardized interfaces. This solves several technical bottlenecks in multimodal human-computer interaction systems and achieves consistency, security, and efficiency across devices.

CN122363494APending Publication Date: 2026-07-10曾裕仁
0 Cites 0 Cited by

Patent Information

Application Number
CN202610262961.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-05
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing multimodal human-computer interaction systems suffer from multimodal timing and consistency issues, difficulty in accumulating and extending personalization and long-term memory across devices, coarse-grained privacy control, lack of auditable dynamic routing mechanisms for end-to-end, edge-to-cloud computing power scheduling, lack of standardized platform interfaces, and insufficient observability and traceability.

Method used

The system adopts a multimodal human-computer interaction platform based on a central coordination layer. The semantic event conversion module converts multimodal inputs into structured semantic event objects, and the personality proxy module generates personalized interactive responses. These responses are provided to the outside world in a standardized interface form through the developer interface module. The system supports privacy classification and multi-tenant policy control, and achieves unified transmission and personalized responses across modules and devices.

Benefits of technology

It achieves consistency, security, and efficiency in cross-application and cross-device interactions, reduces development complexity and cost, and ensures personalized responses and privacy protection across different scenarios.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a multi-modal man-machine interaction platform system and method based on a central coordination layer, and the method comprises the steps: converting multi-modal input data from an edge interaction unit into a structured semantic event object in a unified manner, and enabling the object to at least package a semantic intention, an emotional context, a privacy compliance label and a strategy identifier; the method is used as a unique transmission unit for cross-module and cross-device interaction. In the central coordination layer, emotion understanding, human agency, privacy compliance verification and computing power resource dynamic scheduling capability are integrated, and the central coordination layer is opened to the outside through a standardized interface. Therefore, a third-party developer can directly obtain the platform capability without repeatedly constructing an underlying model, a privacy compliance facility and a cross-end consistency facility, so that the consistency, the safety and the high efficiency of cross-application and cross-device interaction are effectively guaranteed, and the development complexity and the development cost are remarkably reduced.
Need to check novelty before this filing date? Find Prior Art