Road Surface Detection Device Trench Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing road surface detection devices, such as those disclosed in Japanese Laid-Open Patent Publication No. 2017-223511, are unable to accurately detect trenches or ditches on road surfaces, which is crucial for autonomous driving applications.
Innovation Solution
A road surface detection device that uses a LiDAR sensor to scan the road surface and an external environment recognition unit to calculate two-dimensional positional information, projecting detection points onto a coordinate plane to identify changes in the point sequence, thereby detecting trenches with high accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If three-dimensional point data is used for road surface detection, then comprehensive spatial information is obtained, but trench position detection accuracy deteriorates
Solution Approach 1:
The invention extracts two-dimensional positional information from the three-dimensional point data by removing height direction information. This extraction creates a specialized dataset optimized for trench detection while preserving the essential planar position information needed to identify trench locations and configurations.
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
Enables precise detection of trenches and their positions, enhancing the accuracy of road surface configuration information essential for autonomous vehicle navigation.
Implementation Method 1
a sensor configured to detect three-dimensional positional information at individual detection points by scanning a layer that includes the road surface
Data Source
AI summary
A road surface detection device, which detects a road surface of a road on which a host vehicle runs, includes: a sensor configured to detect three-dimensional positional information at individual detection points by scanning a layer that includes the road surface; and an external environment recognition unit configured to calculate two-dimensional positional information in which positional information in a height direction has been removed from the positional information at the individual detection points included in the layer, to calculate, based on the two-dimensional positional information, a point sequence of the individual detection points in a two-dimensional coordinate plane, and to detect a trench existing in the road surface in accordance with a state of change of the point sequence.