Tunnel intelligent robot inspection method

By employing a tunnel intelligent robot inspection method and utilizing the collaborative mechanism of dynamic window alignment, edge computing, and cloud twin center, the latency, rendering, and data synchronization issues of the tunnel digital twin system were resolved, achieving real-time data alignment and system performance optimization.

CN121764072APending Publication Date: 2026-03-31SHANGHAI NATURAL GAS PIPELINE NETWORK CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-12-04
Publication Date
2026-03-31

AI Technical Summary

Technical Problem

Traditional tunnel digital twin systems suffer from high system latency, heavy rendering load, data asynchrony, serious waste of computing power, and inability to self-optimize, resulting in data not being mapped in real time, rendering stuttering, and significant performance fluctuations.

Method used

The tunnel intelligent robot inspection method is adopted. Multi-source sensor data is synchronized through a dynamic window alignment mechanism, edge computing nodes analyze and allocate tasks in real time, and the cloud twin center performs multimodal Kalman filter model fusion and reinforcement learning model optimization to achieve real-time data alignment, edge and cloud load balancing, and adaptive rendering of 3D twin scene.

Benefits of technology

It achieves real-time alignment and compressed transmission of multi-source data, intelligent division of labor and load balancing between the edge and the cloud, and improves the rendering efficiency of 3D twin models and real-time monitoring and dynamic optimization of system performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121764072A_ABST
    Figure CN121764072A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of tunnel inspection, and provides an intelligent robot inspection method for a tunnel, and the method comprises the steps: collecting and uploading multi-source sensing data through an inspection robot; the edge computing node analyzes data in real time, monitors performance indexes, calculates a comprehensive score through a performance scoring function, and dynamically allocates tasks to the high-performance node; the cloud twinning center receives the processed data, fuses multi-source motion state data, and performs three-dimensional twinning scene rendering and state synchronization; collecting system operation performance indexes in real time, inputting the system operation performance indexes into the reinforcement learning model to output action decisions, and dynamically adjusting rendering parameters; and finally, displaying the tunnel state on a visual platform and triggering an abnormal alarm. According to the application, data acquisition real-time alignment and compression transmission, intelligent division of labor and load balancing of the edge and the cloud, adaptive rendering of the three-dimensional twin model, multi-source data fusion precision improvement, and real-time monitoring and dynamic optimization of system operation performance are realized.
Need to check novelty before this filing date? Find Prior Art