Park lane line sensing method and system based on multi-source data fusion and prior map constraint

By using multi-source data fusion and prior map constraints, the problem of unstable lane line perception in the park was solved, and high-precision lane line detection was achieved in complex environments, thus improving the safety and robustness of autonomous driving.

CN122336482APending Publication Date: 2026-07-03DONGFENG COMML VEHICLE CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610334393.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-18
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Lane line perception in the park is unstable in complex environments. Existing technologies have difficulty accurately detecting lane lines under conditions such as changes in lighting, wear and tear of lane markings, and occlusion. Furthermore, high-precision map solutions lack adaptability, posing safety hazards.

Method used

By fusing multi-source data and constraining prior maps, images, vehicle motion states, and positioning data are collected, various visual features are extracted and fused, lane line selection and model fitting are performed in conjunction with high-precision map information, and kinematic models are used for prediction and tracking to construct a confidence assessment system.

Benefits of technology

It significantly improves the robustness and accuracy of lane detection in the park, reduces the false detection and missed detection rates, ensures perception continuity and adaptability, and enhances the safety and reliability of autonomous driving.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a park lane line sensing method and system based on multi-source data fusion and prior map constraint, and relates to the technical field of maps and positioning navigation, and the method comprises the steps: collecting multi-source data, and carrying out the preprocessing of an image; color, edge and semantic features are extracted for weighted fusion, and a candidate lane line thermodynamic diagram is generated; obtaining prior information of a high-precision map, projecting the prior information to an image coordinate system, and performing spatial screening on candidate points; performing adaptive model fitting on the effective point set to obtain a lane line observation value; kalman filtering tracking and prediction are carried out in combination with a vehicle kinematic model; and calculating the confidence coefficient by integrating the multi-dimensional indexes and carrying out grading processing. According to the method, the detection robustness in a complex environment is improved through multi-feature fusion, flexible constraint is carried out by utilizing map prior to eliminate noise, the continuity during shielding is guaranteed by introducing motion tracking, abnormal response is realized through confidence coefficient evaluation, and the perception problem caused by unclear park lane line features and changeable environments is effectively solved.
Need to check novelty before this filing date? Find Prior Art