3D Lane Line Labeling from Vehicle Motion Trajectories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional two-dimensional lane lines are ineffective for accurate tracking and control in automatic driving, especially on non-flat roads, necessitating the development of efficient methods for obtaining 3D lane lines to ensure safe and effective vehicle navigation.
Innovation Solution
A lane line labeling method that utilizes sensors and a convolutional neural network to acquire and process images, combined with GNSS and IMU data, to calculate projected pixel coordinates and determine 3D lane lines through camera coordinates, enabling automatic lane line alignment without manual labeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual labeling is used to obtain 3D lane line data, then labeling accuracy can be ensured, but labeling efficiency is very low and time-consuming
Solution Approach 1:
The system uses the vehicle's own motion trajectory data to automatically label lane lines, making the labeling process self-service without requiring external manual intervention. The vehicle's GPS and IMU data serve as the labeling reference, eliminating the need for manual 3D lane line annotation while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical manual labeling process with an automated computational system. Instead of human operators manually annotating lane lines, the system uses algorithmic processing of motion trajectory data, image data, and sensor data to automatically generate 3D lane line labels, significantly improving efficiency.
2Device complexity
If traditional two-dimensional lane lines are used, then the system is simple to implement, but tracking and control performance deteriorates on non-flat roads
Solution Approach 1:
The patent transitions from 2D lane line representation to 3D lane line representation by incorporating elevation information from motion trajectory data. This dimensional upgrade allows the system to accurately represent lane lines on non-flat roads, improving tracking and control reliability while maintaining reasonable system complexity through efficient coordinate transformation algorithms.
3Measurement precision
If 3D lane line data is obtained through manual methods, then accurate navigation can be achieved, but the process is time-consuming and inefficient
Solution Approach 1:
The system performs preliminary action by collecting and storing motion trajectory data, image data, and sensor data during normal vehicle operation. This pre-collected data is then used for automatic lane line labeling, eliminating the need for time-consuming manual labeling processes while maintaining navigation accuracy.
Solution Approach 2:
The patent creates a digital copy of the physical lane lines by processing sensor data and motion trajectory information. This digital replication captures the 3D characteristics of lane lines without requiring physical measurement or manual labeling, achieving accurate navigation data generation efficiently.
Data Source
AI summary
A lane line labeling method applied to an electronic device is provided. In the method, the electronic device acquires a target image corresponding to a target lane. Motion trajectory points of a target vehicle driving on the target lane are obtained. The electronic device determines projected pixel coordinates of the motion trajectory points on the target image, and determines target pixel coordinates corresponding to target lane lines on the target lane based on the projected pixel coordinates. Once target camera coordinates corresponding to the target pixel coordinates are obtained, the electronic device labels the target lane lines according to the target camera coordinates.