Automatic driving scene mining method, system and terminal device

By identifying and processing the types of vehicle driving data, and using rule mining and deep learning methods to generate autonomous driving scenarios, the problem of inaccurate existing tests is solved, and more accurate and universal autonomous driving tests are achieved.

CN122309333APending Publication Date: 2026-06-30HAOMO TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HAOMO TECH CO LTD
Filing Date
2024-12-27
Publication Date
2026-06-30

AI Technical Summary

Technical Problem

The existing autonomous driving scenario testing is not comprehensive or standardized enough, resulting in inaccurate test results and an inability to effectively assess the safety and reliability of autonomous vehicles.

Method used

By acquiring vehicle driving data and identifying its type as relational or non-relational, different mining methods are used to extract scene information and generate autonomous driving scenarios. For relational data, rule mining methods are used, based on formula calculation and signal analysis; for non-relational data, deep learning algorithms are used to process it and construct detailed driving environment and behavioral information.

Benefits of technology

The generated autonomous driving scenarios are closer to real-world driving scenarios, improving the accuracy and versatility of the tests and enhancing the safety and reliability of autonomous driving systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122309333A_ABST
    Figure CN122309333A_ABST
Patent Text Reader

Abstract

This application applies to the field of autonomous driving technology, providing a method, system, and terminal device for mining autonomous driving scenarios. The method includes: acquiring vehicle driving data and identifying the type of the vehicle driving data, which includes relational data and non-relational data; if the vehicle driving data is relational data, obtaining first scenario information based on the relational data; if the vehicle driving data is non-relational data, obtaining second scenario information based on the non-relational data; and generating an autonomous driving scenario based on the first scenario information and the second scenario information. Through the above method, a general autonomous driving scenario can be obtained.
Need to check novelty before this filing date? Find Prior Art