In-Vehicle Sensor Axial Deviation Estimation Using Road Surface Planes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting axial deviation of in-vehicle sensors based on white line detection are inaccurate when one or both white lines are not detected, the lines are not straight, or the vehicle is not traveling straight.
Innovation Solution
An axial deviation estimation apparatus that includes a road surface recognition unit using a laser radar to recognize the road surface, a plane creation unit to create a virtual plane based on this recognition, and an estimation unit to calculate the axial deviation by determining the inclination of this virtual plane relative to a reference plane of the laser radar.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If white line detection method is used to estimate axial deviation, then the system structure remains simple, but measurement precision deteriorates when white lines are not detected or vehicle does not travel straight
Solution Approach 1:
The patent replaces the optical/image-based white line detection method with a laser radar-based measurement system. The laser radar emits laser beams to actively measure the road surface position, substituting the passive optical detection method. This enables precise axial deviation estimation through direct distance measurement rather than image processing, resolving the contradiction between measurement precision and device complexity by using a more reliable sensing modality.
Solution Approach 2:
The patent introduces a virtual plane as an intermediary computational construct. The laser radar measures actual road surface positions, which are then used to calculate a virtual plane representing the ideal road surface. By comparing the actual plane with the virtual plane, the system estimates axial deviation. This intermediary approach allows the system to handle various road conditions and vehicle states while maintaining measurement precision.
2Reliability
If white line detection is used, then ease of operation is maintained, but reliability deteriorates under various driving conditions
Solution Approach 1:
The patent replaces the complex image processing and white line detection algorithms with direct laser ranging measurements. The laser radar provides reliable distance measurements to the road surface regardless of lighting conditions, lane markings quality, or vehicle trajectory, thereby improving reliability while simplifying the operational requirements.
Solution Approach 2:
The laser radar system performs self-calibration and automatic reference establishment by measuring the road surface and computing the virtual plane. The system autonomously adapts to different road conditions and vehicle states without requiring manual intervention or complex detection algorithms, improving both reliability and ease of operation.
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 allows for accurate estimation of axial deviation even when traditional detection methods fail, ensuring reliable sensor alignment and vehicle control systems function effectively.
Implementation Method 1
a road surface recognition unit configured to recognize a road surface by a laser radar installed in a vehicle
Data Source
AI summary
An axial deviation estimation apparatus of an in-vehicle sensor includes a road surface recognition unit configured to recognize a road surface by a laser radar installed in a vehicle, an estimation unit configured to estimate an axial deviation amount of the in-vehicle sensor mounted on the vehicle, and a plane creation unit configured to create a virtual plane of the road surface based on a recognition result of the road surface recognition unit. The estimation unit estimates the axial deviation amount based on an inclination of the virtual plane with respect to a reference plane of the laser radar.