LIDAR Road Surface Mapping for Slipperiness Estimation Ahead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing LIDAR sensor systems for road condition estimation in vehicles are limited by their reliance on absolute reflectance indices, which are inconsistent and non-generalizable due to factors like distance from the sensor and light scattering, making it difficult to accurately predict slippery road conditions and provide a holistic view of the road ahead.
Innovation Solution
The method involves segmenting the road surface into MxN patches and using LIDAR point clouds to statistically evaluate relative position, feature elevation, and scaled reflectance indices, enabling the determination of slipperiness probability for each patch, which is then used to alert drivers and enable/disable autonomous driving functionalities, and report road conditions to a cloud server.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If LIDAR sensors are mounted close to the ground to enhance visibility for water depth detection, then visibility of road surface conditions is improved, but the ability to detect road conditions at long range is limited
Solution Approach 1:
The patent transitions from 1-D reflectance signals along wheel paths to a 2-D grid of MxN patches covering the road surface ahead. This dimensional expansion allows the system to detect road conditions across a broader area while maintaining long-range detection capability, as each patch provides localized information that collectively covers the entire road surface in front of the vehicle.
Solution Approach 2:
The road surface is divided into MxN patches, where M represents lateral patches and N represents longitudinal patches. This segmentation allows the system to analyze road conditions in discrete zones, enabling both close-range detailed detection and long-range overview by processing multiple patches at different distances simultaneously.
2Length of stationary object
If LIDAR sensors are mounted at higher positions to detect road conditions at long range, then detection range is improved, but the ability to track reflectance along wheel contact paths is reduced
Solution Approach 1:
By implementing an MxN patch grid that covers both lateral and longitudinal dimensions of the road surface, the system recovers the wheel path information even when the LIDAR is mounted at a higher position. The grid structure ensures that patches along the expected wheel contact paths are still captured and analyzed, maintaining the ability to detect reflectance variations critical for identifying slippery conditions.
Solution Approach 2:
The MxN patch system serves multiple functions simultaneously: it provides long-range detection capability, covers the entire road surface area including wheel paths, and enables both localized detailed analysis and global road condition assessment. This multi-functional approach eliminates the trade-off between detection range and visibility area.
3Device complexity
If absolute reflectance indices are used for road condition prediction, then simplicity of measurement is maintained, but reliability of slipperiness detection deteriorates due to distance and light scattering variations
Solution Approach 1:
The patent applies local quality by analyzing reflectance characteristics within each individual patch rather than using a single absolute reflectance value for the entire road surface. Each patch's reflectance is evaluated in the context of its specific location, distance, and local conditions, making the measurement reliable despite variations in distance and light scattering. This localized approach allows the system to account for environmental variations while maintaining measurement simplicity.
4Device complexity
If 1-D reflectance signals along wheel paths are used, then simplicity of analysis is maintained, but holistic view of entire road condition is lost
Solution Approach 1:
The road surface is segmented into MxN patches, transforming the analysis from 1-D signals to a 2-D grid structure. This segmentation preserves the simplicity of analyzing discrete signals while capturing holistic road condition information, as each patch provides independent data about its local condition. The grid structure naturally organizes the information from all patches to form a complete picture of the road surface ahead.
Solution Approach 2:
By adding the lateral dimension (M patches) to the existing longitudinal dimension (N patches), the system expands from 1-D to 2-D analysis. This dimensional change enables comprehensive coverage of the entire road surface while maintaining manageable analysis complexity through the structured grid format. The MxN organization allows efficient processing and provides a complete spatial map of road conditions.
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 provides a more reliable and scalable method for predicting road conditions, enhancing safety by enabling accurate slipperiness detection and path planning, even under varying lighting conditions, and improving the reliability of autonomous driving systems.
Implementation Method 1
LIDAR sensors are generally capable of detecting the road surface in terms of relative road surface location (x and y coordinates), road surface height (z coordinate), and reflectance (r) reliably up to 70-100m ahead
Implementation Method 2
LIDAR sensors have been used to track reflectance on a road surface along two straight-line paths
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for estimating road conditions ahead of a vehicle, including: a LIDAR sensor operable for generating a LIDAR point cloud (16); a processor executing a road condition estimation algorithm stored in a memory, the road condition estimation algorithm performing the steps including: detecting a ground plane or drivable surface in the LIDAR point cloud; superimposing an MxN matrix on at least a portion of the LIDAR point cloud; for each patch of the LIDAR point cloud defined by the MxN matrix, statistically evaluating a relative position, a feature elevation, and a scaled reflectance index; and, from the statistically evaluated relative position, feature elevation, and scaled reflectance index, determining a slipperiness probability for each patch of the LIDAR point cloud; and a vehicle control system operable for, based on the determined slipperiness probability for each patch of the LIDAR point cloud, affecting an operation of the vehicle.