LiDAR White Line Position Estimation Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of estimating a vehicle's position using white lines detected by LiDAR varies significantly due to differences in data quantity and quality, such as continuous vs. broken lines and paint deterioration, leading to inconsistent detection accuracy.
Innovation Solution
A measurement device and method that acquire positional information and point group data from external sensors, calculating and outputting the reliability of the vehicle's position based on the number and distribution of points within a predetermined range, thereby adjusting the use of positional information to prevent accuracy deterioration caused by low reliability data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If white lines are used for position estimation, then position estimation can be performed, but detection accuracy varies significantly due to differences in data quantity and quality
Solution Approach 1:
The patent changes the parameter of reliability assessment by introducing a quantitative evaluation metric that measures the number and distribution of LiDAR detection points along white lines. This parameter change allows the system to dynamically assess whether detected white lines meet the required detection accuracy threshold, thereby resolving the inconsistency in detection reliability caused by varying data quantity and quality.
2Quantity of substance
If LiDAR data is used to detect white lines, then position information can be obtained, but the amount of data varies depending on white line types and paint deterioration
Solution Approach 1:
The patent introduces a dynamic reliability evaluation mechanism that adaptively assesses the quality of LiDAR detection data based on the actual number of detected points and their distribution along the white line. Instead of using a fixed data quantity threshold, the system dynamically determines whether the detected white line is reliable for position estimation, allowing it to handle variations in data quantity caused by different white line types and paint conditions.
3Productivity
If all detected white lines are used for position estimation, then more data is available, but accuracy deteriorates when low reliability data is included
Solution Approach 1:
The patent applies local quality control by evaluating and filtering individual white line detections based on their specific reliability metrics rather than uniformly processing all detected lines. The system assesses each white line's point distribution and density locally, accepting only those that meet the reliability threshold for position estimation. This selective approach ensures that only high-quality local detections contribute to the overall position estimation, maintaining accuracy while preserving efficiency.
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
This approach enhances the accuracy of vehicle position estimation by weighting the reliability of detected white lines, ensuring that only high-reliability data is used for precise positioning, thus stabilizing the estimation process regardless of data quality variations.
Implementation Method 1
detecting white lines based on its reflectivity
Data Source
AI summary
A measurement device includes a processor coupled to a memory storing instructions. The processor is configured to acquire positional information of a measurement object stored in a storage unit, acquire point group information of points indicating a surrounding feature acquired by an external sensor, output reliability of the positional information of the measurement object indicating a center position of the measurement object existing in a predetermined range based on the point group information of points existing in the predetermined range, the predetermined range being determined based on a position of a movable body, and estimate a movable body position based on the positional information of the measurement object and the center position of the measurement object, wherein the processor determines the center position by calculating an average value of the point group information existing in the predetermined range.