Autonomous driving method of locomotive

The autonomous driving method employs an AI model to efficiently generate and transmit control commands for locomotives, addressing time-consuming manual modeling and condition variability, enhancing precision and adaptability.

WO2026127209A1 Publication Date: 2026-06-18POSCO HLDG INC
0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
POSCO HLDG INC
Filing Date
2025-01-10
Publication Date
2026-06-18

AI Technical Summary

⚠Technical Problem

Existing methods for autonomous locomotive control require significant time for manual modeling and optimization, and yield varying results due to locomotive condition changes, necessitating frequent re-modeling.

⚗Method used

An autonomous driving method utilizing a multi-layer artificial intelligence model to process sensor data and image information for real-time control value generation, including control commands like neutral, reverse, and braking, with constraint-based transmission.

🎯Benefits of technology

Minimizes control optimization time and reduces errors by automating control modeling, ensuring precise control values and adaptability to locomotive conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

An autonomous driving method of an autonomous driving system of a locomotive comprises the steps of: acquiring a surrounding image and state information from a camera and a plurality of sensors installed in the locomotive; extracting a region-of-interest image of a track area on the basis of the surrounding image; extracting image feature information by inputting the region-of-interest image to first layers of an artificial intelligence model; extracting driving information including a target speed, a remaining distance to a destination, and a plurality of waypoints on the basis of the current location of the locomotive; extracting state and driving feature information by inputting the state information and the driving information to second layers of the artificial intelligence model; and outputting any one control value among a plurality of control values including a neutral control value, a backward control value, a genetic control value, a release control value, and a braking control value by inputting the image feature information and the state and driving feature information to third layers of the artificial intelligence model.
Need to check novelty before this filing date? Find Prior Art