Quality evaluation driven offshore target multi-modal information collection agent system and method

CN122284560APending Publication Date: 2026-06-26NAVAL AVIATION UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NAVAL AVIATION UNIV
Filing Date
2026-06-01
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing multimodal information acquisition systems struggle to balance acquisition quality and execution stability in long-term operational scenarios targeting maritime targets, and are prone to system instability when acquisition anomalies occur.

Method used

A quality assessment-driven execution stability control module is introduced. Through a task interaction and constraint module, a multimodal information acquisition and execution module, a multimodal parsing and alignment module, and a data storage module, the quality assessment results are used as endogenous control signals. Combined with attribution mapping and differential execution adjustment, dynamic adjustment of the acquired execution parameters is achieved.

Benefits of technology

Without altering the overall data acquisition process structure, the quality and stability of multimodal information acquisition and the continuity of system operation were improved, avoiding the accumulation of low-quality data and system oscillations, and ensuring long-term operational stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122284560A_ABST
    Figure CN122284560A_ABST
Patent Text Reader

Abstract

This invention provides a quality assessment-driven intelligent agent system and method for acquiring multimodal information about maritime targets, belonging to the field of multimodal information acquisition. The system includes: a task interaction and constraint module, a central orchestration agent module for generating acquisition execution plans and scheduling tasks; a multimodal information acquisition execution module, a multimodal parsing and alignment module for parsing, structuring, and aligning data across modalities; an execution stability control module for converting quality changes in structured multimodal results into control signals for the acquisition execution process and differentially adjusting the acquisition execution parameter vector under constraints; and a data storage and traceability module. By introducing a quality assessment-driven execution stability control module, the quality changes in acquisition results are used as endogenous control signals, and combined with attribution mapping, parameter subspace constraints, and differential execution adjustment mechanisms, long-term stable operation of the multimodal information acquisition process is achieved.
Need to check novelty before this filing date? Find Prior Art