Localized LiDAR Surface Fitting for Road Profile Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for estimating road surfaces in autonomous vehicles lack accuracy, particularly due to limitations in LiDAR and RADAR technologies, leading to poor handling on uneven surfaces and potential vehicle instability.
Innovation Solution
A method using localized surface fitting and bias correction of LiDAR data, combined with stereo imaging, to estimate a 3D road surface profile, which involves ego-motion compensation, accumulation of LiDAR data, and nonlinear optimization to improve accuracy and robustness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR is used to create detailed 3D representations of the environment, then measurement precision is improved at close range, but data sparsity increases at greater distances
Solution Approach 1:
The patent combines LiDAR data with RADAR data and camera images to create a fused representation of the environment. This multi-sensor fusion approach compensates for LiDAR's data sparsity at long ranges by incorporating complementary information from other sensors, thereby maintaining measurement precision without relying solely on dense LiDAR point clouds.
Solution Approach 2:
The system employs multiple sensor types (LiDAR, RADAR, cameras) that serve multiple functions. LiDAR provides detailed 3D structure at close range, RADAR extends detection to long ranges in various weather conditions, and cameras provide high-resolution visual information. This multi-functional sensor suite ensures comprehensive environmental perception across different operating conditions.
2Length of stationary object
If RADAR is used to detect objects over long ranges, then detection range is improved, but measurement precision deteriorates
Solution Approach 1:
The patent fuses RADAR data with LiDAR and camera data to achieve both long detection range and high measurement precision. RADAR's strength in long-range detection is combined with LiDAR's and camera's strength in precise surface characterization, creating a synergistic system that overcomes the individual limitations of each sensor.
Solution Approach 2:
The system uses camera images as an intermediary to bridge the precision gap in long-range RADAR measurements. Camera data provides high-resolution visual information that complements RADAR's long-range detection capability, allowing the system to achieve both extended detection range and accurate surface estimation.
3Illumination intensity
If camera-only solutions are used to capture high-resolution images, then image resolution is improved, but depth estimation precision deteriorates
Solution Approach 1:
The patent combines monocular camera data with depth information from LiDAR and stereo vision to achieve both high image resolution and accurate depth estimation. The camera provides detailed visual texture and color information, while integrated depth sensors and stereo processing supply precise distance measurements, creating a comprehensive 3D environmental model.
Solution Approach 2:
The system transitions from 2D camera images to 3D environmental understanding by integrating depth information from multiple sources. Stereo vision computes depth from disparities between left and right camera images, while LiDAR directly measures distances, adding the third dimension to the camera's 2D visual data and enabling accurate 3D surface reconstruction.
4Measurement precision
If LiDAR data is accumulated and processed using nonlinear optimization, then surface estimation accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the environment into discrete surface patches and processes each patch independently through nonlinear optimization. This segmentation approach allows the complex computational task to be divided into manageable sub-problems, reducing overall computational complexity while maintaining high surface estimation accuracy for each local region.
Solution Approach 2:
The system performs preliminary processing of LiDAR data including ego-motion compensation and bias correction before applying nonlinear optimization. By pre-processing the data to remove systematic errors and compensate for vehicle movement, the subsequent optimization process becomes more efficient and converges faster, reducing the overall computational burden.
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 the accuracy and robustness of road surface estimation, enabling safe and comfortable autonomous vehicle operation by accurately detecting obstacles and adapting suspension systems to road conditions.
Implementation Method 1
LiDAR, which uses laser pulses to create detailed 3D representations of the environment
Implementation Method 2
RADAR, which uses radio waves to detect objects and measure distances
Data Source
AI summary
Embodiments of the present disclosure relate to ground surface estimation using localized surface fitting. A three-dimensional (3D) surface structure (e.g., a road surface profile) may be estimated using a nonlinear optimization to fit height values to (e.g., accumulated, bias-corrected) LiDAR detections (e.g., sampled in localized regions along one or more predicted trajectories). For example, LiDAR data (e.g., detected 3D point clouds) may be ego-motion compensated, corrected for measurement bias, accumulated, and sampled along one or more predicted trajectories, and the height of each trajectory point may be fitted to the heights of the corresponding sampled points using a nonlinear optimization. As such, the resulting road surface profile (e.g., modeled along the wheel track(s)) may be provided to an adaptive suspension control system to modulate the damping characteristic of the suspension system to counteract indentations (e.g., potholes) or protrusions (e.g., speed bumps) represented in the road surface profile.