Slope-Based Lane Marking Discretization for Map-less Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Map-less and camera-based Level 3 autonomous driving vehicles face challenges in accurately detecting and discretizing lane markings for motion planning without relying on map data, which affects their ability to navigate effectively.
Innovation Solution
A computer-implemented method that generates a polynomial representing a lane boundary line, discretizes it into points based on the slope at previous points, and uses these discretization points to create a lane reference line for trajectory planning, enabling the vehicle to navigate within the lane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polynomial discretization is performed for lane marking detection in map-less autonomous driving, then the vehicle can generate reference lines for motion planning, but the discretization accuracy may be insufficient for complex curved lanes
Solution Approach 1:
The patent segments the polynomial discretization process into multiple stages: first identifying key control points on the lane marking, then fitting polynomials to these points, and finally discretizing the polynomials at multiple levels (coarse and fine discretization). This segmentation allows accurate representation of complex curved lanes while managing computational complexity through hierarchical processing.
Solution Approach 2:
The patent implements dynamic discretization where the discretization interval is adjusted based on the local curvature of the lane marking. In regions with high curvature, smaller intervals are used to maintain accuracy, while in straight sections, larger intervals are used to reduce computational load. This dynamic adaptation resolves the contradiction between accuracy and complexity.
2Adaptability or versatility
If the vehicle operates without map data, then it gains versatility in unknown environments, but lane marking detection becomes more difficult without reference information
Solution Approach 1:
The system performs self-service by using onboard cameras to detect and reconstruct lane markings autonomously without external map data. The polynomial fitting and discretization processes enable the vehicle to create its own reference framework from raw visual input, making the system self-sufficient in unknown environments while maintaining detection accuracy.
Solution Approach 2:
The patent replaces traditional mechanical or map-based navigation systems with a vision-based detection system. By substituting camera-based polynomial fitting for map-matching approaches, the system achieves versatility in unknown environments while the sophisticated image processing algorithms maintain detection accuracy despite the increased difficulty.
3Manufacturing precision
If discrete points are densely sampled along the polynomial, then the reference line accuracy improves, but the computational load increases
Solution Approach 1:
The patent applies local quality by using non-uniform discretization where points are densely sampled in regions of high curvature and sparsely sampled in straight sections. This approach maintains reference line precision where needed while reducing computational power consumption in areas where high precision is not necessary, effectively resolving the contradiction between accuracy and power usage.
Data Source
AI summary
A computer-implemented method, apparatus, and system for discretizing lane markings and for generating a lane reference line is disclosed. A polynomial defined over an (x,y) coordinate system is received, the polynomial being representative of at least a portion of a lane boundary line. A length of the polynomial is determined. The polynomial is discretized, comprising determining a plurality of discretization points on the polynomial to represent the polynomial, wherein a first discretization point is a first end of the polynomial, wherein subsequent discretization points are determined successively until the polynomial is completely discretized, and wherein each discretization point other than the first discretization point is determined based at least in part on a slope of the polynomial at a previous discretization point. Thereafter, a lane reference line comprising a plurality of points is generated based on the discretized polynomial.