Lane Edge Tracking Curve Fitting for Noisy Road Markings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane detection technologies face challenges in accurately extracting lane edges due to blurred road signs, accumulated snow, and other environmental factors, leading to inaccurate vehicle control and increased risk during autonomous driving.
Innovation Solution
A lane edge extraction method that utilizes tracking edge points from preceding frames, determines observation edge points in current frames, and corrects these points using a combination of vehicle rectangular and polar coordinate systems to fit a lane edge curve, excluding outliers and iteratively updating tracking edge points to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing visual identification is used to provide drivable space information, then the basic function of lane detection is achieved, but the data is noisy and has large errors which raises risks for route planning and control
Solution Approach 1:
The patent applies preliminary action by pre-establishing lane edge curves based on historical tracking data before new data arrives. The system predicts lane edge positions in advance and uses these predictions to guide the extraction and filtering of new observation points, ensuring that data processing starts with a reliable reference framework that reduces the impact of noisy input data
Solution Approach 2:
The patent implements feedback mechanisms by continuously comparing new observation edge points with the established lane edge curve and using the deviations to adjust and refine the curve. The system feeds back the fitted curve results to guide subsequent data collection and processing, creating a closed-loop system that progressively improves accuracy while filtering out noisy data through iterative refinement
2Adaptability or versatility
If traditional lane detection methods are used under adverse conditions (blurred road signs, water, snow), then basic detection functionality is maintained, but accuracy and stability deteriorate significantly
Solution Approach 1:
The patent applies dynamics by implementing a dynamic lane edge tracking system that continuously adapts to changing road conditions. The system dynamically updates the lane edge curve as new observation data becomes available, allowing it to adapt to blurred signs, water, and snow conditions while maintaining accuracy through continuous refinement rather than relying on static detection thresholds
Solution Approach 2:
The patent uses parameter changes by transforming the lane edge representation from discrete point detection to continuous curve fitting. By changing the mathematical representation parameters from simple edge points to fitted curves with multiple degrees of freedom, the system can better handle adverse conditions through smoothing and interpolation, reducing the impact of localized noise from blurred signs or weather effects
3Speed
If lane edge data is directly used for route planning and control without additional processing, then real-time responsiveness is maintained, but safety risks increase due to noisy and error-prone data
Solution Approach 1:
The patent applies preliminary action by pre-fitting lane edge curves and establishing reference frameworks before critical decision-making moments. The system performs curve fitting and data refinement in advance, so that when route planning and control decisions are needed, reliable processed data is already available, maintaining real-time responsiveness without sacrificing safety
Solution Approach 2:
The patent introduces an intermediary processing layer (the lane edge curve fitting and filtering system) between raw observation data and the route planning/control systems. This intermediary transforms noisy raw data into refined curve representations that serve as a reliable bridge, enabling real-time processing while filtering out errors before they reach the critical decision-making systems
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
This application relates to a lane edge extraction method and apparatus, an autonomous driving system, a vehicle, and a storage medium. The lane edge extraction method includes the steps of: receiving tracking edge points, about lane edges, of an immediately preceding frame of an edge image sequence; determining observation edge points, about the lane edges, of a current frame of the edge image sequence; continuing and correcting the tracking edge points of the immediately preceding frame based on the observation edge points of the current frame, to obtain temporary tracking edge points of the current frame; fitting a lane edge curve based on the temporary tracking edge points; and excluding outliers from the temporary tracking edge points based on the lane edge curve, to form tracking edge points of the current frame. The method can improve the stability and accuracy of lane edge extraction.