Lane Keeping Correction Unit for Misrecognized Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lane keeping assistant systems face instability and reduced reliability due to misrecognized lane information, especially under external and environmental factors, as they fail to accurately consider vehicle heading angle, lane curvature, and rate of curvature changes.
Innovation Solution
A vehicle steering control apparatus that includes a correction unit to estimate and correct misrecognized lane information using self-feedback, determining a permissible recognition range and calculating average lane widths to improve lane information accuracy, thereby enhancing lane keeping reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lane information is measured from a camera, then lane keeping assistance can be provided, but misrecognition occurs due to external factors (guardrails, interconnections, tunnels) and environmental factors (snow, rain, backlight)
Solution Approach 1:
The system uses feedback mechanisms by comparing currently measured lane information with previously stored lane information. When a discrepancy is detected, the system determines whether this represents misrecognition by comparing with historical data, thereby improving reliability while maintaining measurement accuracy.
Solution Approach 2:
The system performs preliminary actions by storing lane information in advance during normal driving conditions. This pre-stored information serves as a reference for later comparison when measurement accuracy may be compromised by external or environmental factors.
2Reliability
If multiple filters are used to estimate lane information, then noise compensation is achieved, but accuracy is lowered because lane width is used as an arbitrary predetermined value instead of actual road information
Solution Approach 1:
The system changes the parameter used for lane estimation from arbitrary predetermined lane width values to actual measured lane information from previous time points. This allows the estimation to adapt to real road conditions rather than relying on fixed assumptions, thereby improving accuracy while maintaining stability through temporal smoothing.
3Reliability
If lane misrecognition is determined based on camera reliability and lane distance variation, then some correction is possible, but accuracy is limited because vehicle heading angle, lane curvature, and rate of curvature change are not considered
Solution Approach 1:
The system enhances the misrecognition detection function by integrating multiple parameters beyond just camera reliability and lane distance variation. It incorporates vehicle heading angle, lane curvature, and rate of curvature change into a comprehensive detection mechanism that provides more accurate and reliable lane information correction.
Data Source
AI summary
Provided are related to a vehicle steering control apparatus and a method for lane keeping, in which the performance of lane keeping support is improved by correcting lane information of a camera. The vehicle steering control apparatus for lane keeping includes a correction unit which estimates lane information to be subsequently input using self-feedback lane information, and corrects lane information determined in a misrecognized state when present input lane information is determined in the misrecognized state using the estimated lane information, and a steering controller configured to control steering of a vehicle in consideration of the lane information corrected by the correction unit.