LiDAR Track Reliability Evaluation Under Contaminated Point Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional LiDAR systems often provide inaccurate point data due to contamination, leading to unreliable tracks that can cause errors in vehicle driving control, necessitating a method to determine the reliability of LiDAR tracks.
Innovation Solution
A method and system that assess the reliability of LiDAR tracks by evaluating tracking input information and function processing reliability, applying weights to error information and predetermined features such as point number, position, contours, shape, occlusion, and viewing angle, and processing reliability of functions like speed extraction and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LiDAR point data is used for object detection, then object detection capability is provided, but accuracy deteriorates due to contamination and coverage of points
Solution Approach 1:
The patent segments the LiDAR tracking system into multiple evaluation components: tracking input information reliability (evaluating point cloud data quality) and function processing reliability (evaluating object detection algorithm performance). By dividing the reliability assessment into these segments, the system can identify specific sources of error and apply targeted corrections to improve overall track reliability despite contamination issues.
Solution Approach 2:
The patent changes parameters by introducing weight coefficients that dynamically adjust the contribution of different reliability factors. The processor applies weights to tracking input information reliability and function processing reliability based on reference values, allowing the system to adapt to varying contamination levels and maintain accurate object detection even when point data quality deteriorates.
2Reliability
If conventional object detection systems generate LiDAR tracks, then object detection is performed, but reliability information is lost leading to control errors
Solution Approach 1:
The patent implements feedback by having the processor continuously evaluate reliability of tracking input information and function processing, then use this reliability information to adjust subsequent tracking and detection operations. The system feeds back reliability scores and weight coefficients to improve the quality of LiDAR tracks, preventing control errors by maintaining awareness of data quality throughout the detection process.
3Reliability
If reliability evaluation of tracking input information and function processing is performed, then reliable LiDAR tracks are identified, but system complexity increases
Solution Approach 1:
The patent applies universality by designing a processor that performs multiple functions: it processes LiDAR point cloud data, evaluates tracking input reliability, assesses function processing reliability, and generates weighted reliability scores. This multi-functional approach consolidates complexity into a single processing unit rather than requiring separate systems for each function, making the reliability determination feasible within existing LiDAR system architectures.
Data Source
AI summary
A method for determining reliability of a LiDAR track includes determining reliability of tracking input information for tracking a target object represented by the LiDAR track based on the reliability of each predetermined features related to the LiDAR track; determining reliability of function processing of a system based on processing reliability of each of predetermined functions of the system for tracking the target object; and determining reliability of the LiDAR track based on the reliability of the tracking input information and the reliability of the function processing and output reliability information.