Environmental Target Feature Points for Large Vehicle Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately representing large environmental targets, such as trucks, using central point attributes, leading to poor lateral and longitudinal control, which can result in traffic accidents.
Innovation Solution
A method for extracting feature points of environmental targets, including laterally closest, farthest, longitudinally closest, and intersection points with lane lines, to provide more comprehensive target information for decision-making systems, enhancing control accuracy and safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If central point attributes are used to represent large environmental targets, then the representation is simple, but the lateral and longitudinal control accuracy deteriorates
Solution Approach 1:
The patent segments the target representation from a single central point to multiple feature points (including lateral closest point, lateral farthest point, longitudinal closest point, and intersection points with lane lines). This segmentation allows the system to capture the spatial extent and geometric characteristics of large targets, thereby improving control accuracy while maintaining manageable complexity through structured feature selection.
2Measurement precision
If comprehensive target information is extracted, then control accuracy improves, but decision-making computation complexity increases
Solution Approach 1:
The patent extracts only the most relevant feature points from the complete target geometry - specifically the lateral closest point, lateral farthest point, longitudinal closest point, and intersection points with lane lines. This selective extraction provides sufficient information for accurate control while filtering out redundant data, thus balancing computational efficiency with control precision.
Solution Approach 2:
The patent applies local quality by assigning different semantic meanings and importance to different feature points based on their geometric locations. For example, the lateral closest point is critical for lane keeping decisions, while intersection points with lane lines are crucial for lane change assessments. This localized importance assignment optimizes computational resources by focusing processing on the most decision-critical features.
Data Source
AI summary
A method and apparatus used for extracting a feature point of an environmental target, which relate to the field of vehicles. The method comprises: acquiring the position of each corner point of an environmental target; and determining a feature point of the environmental target according to the position of each corner point of the environmental target. Thus, the problem wherein target identification is not precise may be solved, which is particularly suitable for the identification of a large target.