Driving scheduling method and device based on large model agent

By applying a driving scheduling method based on large-model intelligent bodies in the rail transit network command center, the problem of driving scheduling in the existing technology is solved, intelligent assistance for driving scheduling is realized, and the level of intelligent supervision is improved.

CN119398460AActive Publication Date: 2025-02-07BEIJING BII ERG TRANSPORTATION TECH CO LTD +1
View PDF 4 Cites 1 Cited by

Patent Information

Application Number
CN202510000168.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-01-02
Publication Date
2025-02-07
Estimated Expiration
2045-01-02

AI Technical Summary

Technical Problem

The existing technology cannot achieve flexibility and intelligence in driving scheduling in the rail transit network command center, and strongly relies on human intervention and understanding.

Method used

Using the driving scheduling method based on large-model agents, the driving scheduling agent is defined through the Autogen environment framework, the vector database, Kafka distributed message queue, query tool set and execution tool set are configured to build a reflection agent and combine it with driving scheduling. Agents are associated to enable real-time perception and feedback.

Benefits of technology

The intelligent driving scheduling of the rail transit network command center has been realized, assisted in completing the tasks of driving scheduling personnel, and improved the intelligent driving supervision of the urban rail network.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119398460A_ABST
    Figure CN119398460A_ABST
Patent Text Reader

Abstract

The invention discloses a driving scheduling method and device based on a large model agent, and relates to the technical field of driving scheduling, and the method comprises the steps: carrying out the role definition and optimization processing of a driving scheduling agent of a network command center based on an Autogen environment framework, and obtaining an optimized driving scheduling agent; constructing a reflection agent; associating the optimized driving scheduling agent with the reflection agent to obtain a final driving scheduling agent; according to the final driving scheduling intelligent agent, the current line network train operation condition is sensed in real time; generating a feedback result when the running condition of the current net train is abnormal; the feedback result comprises the steps of informing corresponding driving scheduling personnel and generating an emergency processing scheme. According to the invention, rail transit network command center driving scheduling intelligence can be realized, daily work of center driving scheduling personnel can be completed in an auxiliary manner, and the intelligent level of urban rail network driving monitoring is improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Intelligent public transport cloud-brain system based on big data

    CN110705747A

  • Urban rail transit train operation control system oriented to different intelligent levels

    CN111923966A

  • Unmanned equipment autonomous task planning and execution method based on large model agent framework

    CN119090307A

  • Routes optimization system based on railroad and street traffic information

    US20230314148A1

Cited By

  • Multi-agent flexible job shop scheduling method and device

    CN120562727A