LiDAR Object Sensing for Wet Road Mirror Image Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LiDAR systems fail to accurately distinguish between real objects and mirror images caused by wet road surfaces, leading to reduced accuracy and potential misidentification of objects during adverse weather conditions.
Innovation Solution
An object sensing device that includes a road surface shape estimation unit, a road surface condition estimation unit, and an observation point determination unit to classify LiDAR observation points into real and mirror images by estimating the road surface condition and identifying low observation points relative to the road surface, inverting them to determine mirror images, and grouping these points for accurate object detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR observes the outside world using a pulsed laser beam, then distance measurement and object detection are enabled, but mirror images are generated on wet road surfaces causing false object detection
Solution Approach 1:
The patent segments observation points into two categories: real observation points and mirror image observation points. By classifying observation points based on their vertical position relative to the road surface, the system separates valid detection data from false mirror image data, thereby improving object detection accuracy while eliminating false positives caused by wet road surface reflections
Solution Approach 2:
The patent inverts the coordinate system for observation points below the road surface. By applying coordinate inversion to points that should not exist physically (mirror images), the system identifies and excludes these false observations, resolving the contradiction between maintaining comprehensive observation coverage and eliminating mirror image interference
2Loss of information
If the LiDAR detects all observation points in the detection region, then complete object information is obtained, but mirror images on wet surfaces are also detected as real objects
Solution Approach 1:
The patent performs preliminary classification of observation points before final object detection. By pre-identifying and excluding mirror image observation points based on their position relative to the road surface, the system ensures that only reliable real object data proceeds to the detection stage, maintaining information completeness while guaranteeing detection reliability
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
Enhances object detection accuracy by classifying LiDAR observation points into real and mirror images, reducing misinterpretation and improving recognition accuracy in wet conditions.
Implementation Method 1
The LiDAR measures a distance to an object by emitting a pulsed laser beam and measuring a time interval until reflected light from the object is received
Implementation Method 2
measuring a time interval until reflected light from the object is received
Implementation Method 3
when the road surface becomes wet due to rainfall or snow, the road surface becomes like a mirror that reflects light
Data Source
AI summary
An object of the present invention is to provide an object sensing device that classifies an observation point group output by a LiDAR into a real image and a mirror image when a road surface around an own vehicle is wet, and can use the mirror image for detecting the real image. An object sensing device that detects an object around a vehicle based on a point cloud data of an observation point observed by a LiDAR mounted on the vehicle includes: a road surface shape estimation unit that estimates a shape of a road surface; a road surface condition estimation unit that estimates a dry/wet situation of the road surface; and an observation point determination unit that determines a low observation point observed at a position lower than the estimated road surface by a predetermined amount or more when the road surface is estimated to be in a wet situation. The object is detected by using point cloud data of the observation points other than the low observation point and point cloud data of an inverted observation point obtained by inverting the low observation point with reference to a height of the road surface.