Vehicle Lane Control Using Lane Width Change Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems face mis-recognition of lane lines due to varying road environments, leading to incorrect steering control when lane width increases, particularly in diverged or expanded stretches.
Innovation Solution
A vehicle control system that determines the expected and real lane widths using camera data, recognizing the opposite lane line to maintain control when lane width increases, preventing mis-recognition and ensuring stable driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If steering control is performed using only curvature information of lane lines, then the control system is simple, but mis-recognition occurs due to diversity of driving road environment
Solution Approach 1:
The system uses lane width information as feedback to verify and correct lane line recognition. By continuously monitoring whether the recognized lane width is reasonable (within expected ranges), the system can detect mis-recognition errors and switch to alternative recognition methods, thereby improving reliability without significantly increasing complexity
Solution Approach 2:
The system pre-establishes expected lane width ranges based on road environment characteristics before actual lane line recognition occurs. This preliminary preparation allows the system to quickly validate recognition results and switch to alternative methods when mis-recognition is detected, improving response time and accuracy
2Ease of operation
If the vehicle recognizes the diverged stretch line as the driving lane line, then the vehicle can follow the visible line, but incorrect steering control occurs
Solution Approach 1:
The system dynamically adjusts its lane line recognition strategy based on real-time lane width validation. When the recognized lane width falls outside expected ranges, the system switches from following the visible line to using alternative recognition methods, making the system adaptive to different road conditions while maintaining reliability
Solution Approach 2:
The system introduces lane width information as an intermediary validation layer between lane line detection and steering control. This intermediary check prevents direct steering control based on potentially incorrect lane line recognition, filtering out mis-recognition cases before they affect steering accuracy
3Device complexity
If the vehicle uses only curvature information for steering control, then the control method is simple, but mis-control occurs in stretches with diverged or expanded lanes
Solution Approach 1:
The system replaces pure curvature-based mechanical steering control with a hybrid approach that incorporates lane width validation. By substituting the single-curvature method with a multi-parameter validation system, the patent achieves higher steering precision in complex road environments while maintaining reasonable system complexity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A vehicle control system and method, the method including: recognizing a driving lane ahead of a vehicle by means of a camera included in the vehicle, and determining an expected lane width of the driving lane over time; determining whether a real lane width of the driving lane increases at a time point at which the determined expected lane width increases; and when it is determined that the real lane width increases, controlling driving, based on information on a line, among both lines of the driving lane, positioned in an opposite direction of a direction in which the lane width increases.