LiDAR Static Object Validation for Accurate Vehicle Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Atypical objects such as bushes and trees can degrade the accuracy of precise positioning in autonomous vehicles by being misclassified as static objects, leading to incorrect vehicle positioning.
Innovation Solution
A LiDAR-based object recognition method and apparatus that processes data from environmental sensors, including LiDAR, to determine the validity of unidentified static objects by generating occupancy and class grid maps, clustering, and evaluating the effectiveness of contour segments, thereby distinguishing and excluding atypical objects from static object information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all detected static objects are used for positioning, then more objects are available for map matching, but atypical objects like bushes and trees degrade positioning accuracy
Solution Approach 1:
The patent segments the set of detected static objects into two categories: typical objects (guardrails, curbs, buildings, signs) and atypical objects (bushes, trees). This segmentation allows the system to selectively use only typical objects for positioning, resolving the contradiction by filtering out harmful objects while maintaining a sufficient quantity of valid objects for map matching.
Solution Approach 2:
The patent applies different quality criteria to different objects based on their characteristics. Typical objects with regular geometries and stable positions are selected for positioning, while atypical objects with irregular shapes are excluded. This local quality approach ensures high positioning accuracy by using only objects that meet specific quality standards.
2Measurement precision
If contour validation is applied to filter atypical objects, then positioning accuracy improves, but processing complexity increases
Solution Approach 1:
The patent performs preliminary contour validation on detected static objects before using them for positioning. By pre-filtering objects based on contour characteristics (regularity, geometric properties), the system eliminates atypical objects early in the processing pipeline, preventing them from degrading positioning accuracy while maintaining efficient processing.
Solution Approach 2:
The patent changes the validation parameters from simple object detection to contour-based analysis. By examining contour regularity, geometric properties, and spatial distribution, the system can distinguish typical from atypical objects using quantifiable parameters, achieving accurate filtering without excessive processing complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves the accuracy of object recognition and prevents incorrect vehicle positioning by effectively filtering out atypical objects, enhancing the precision index of recognized static objects for map matching.
Implementation Method 1
A method may comprise receiving surrounding environment data from at least one environmental sensor including a LiDAR sensor
Data Source
AI summary
A LiDAR-based object recognition method may be performed by an apparatus. The LiDAR-based object recognition method comprises obtaining surrounding environment data from at least one environmental sensor including a LiDAR sensor, obtaining object information for each sensor, including information of a LiDAR static object, based on data processing of the surrounding environment data from each sensor, determining validity of at least one unidentified class static object among the at least one LiDAR static object and outputting static object information according to the validity result.