Lane Estimation Using Map Data When White Line Detection Fails
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lane estimation systems struggle to accurately determine the traveling lane of a vehicle under various travel environments, particularly when cameras fail to recognize white lines.
Innovation Solution
A lane estimation apparatus that utilizes an external circumstance detection part and an electronic control unit to estimate the current lane based on external circumstances, calculating a parameter indicating the frequency of the lane traveled over a predetermined section, and generates a map using this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera-based white line recognition is used to estimate the traveling lane, then the system structure remains simple, but the lane estimation accuracy deteriorates when the camera cannot recognize white lines under various travel environments
Solution Approach 1:
The patent introduces map data as an intermediary to bridge the gap between camera detection and lane estimation. When the camera cannot recognize white lines, the system uses map data containing road structure information (branch portions, joining portions, lane configurations) as a mediator to determine the traveling lane, thereby maintaining accuracy across various travel environments
Solution Approach 2:
The system performs preliminary actions by pre-acquiring and storing map data that includes road structure information before lane estimation is needed. This preliminary preparation of map data allows the system to quickly and accurately determine the traveling lane even when real-time camera recognition fails, improving adaptability to various travel environments
2Reliability
If statistical processing of time-based current lane frequency is performed to estimate the traveling lane, then the reliability of lane estimation improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies partial action by performing statistical processing selectively - it estimates the time-based current lane at discrete time intervals rather than continuously, and calculates frequency parameters only for predetermined sections. This partial statistical processing maintains reliability while reducing unnecessary computational overhead and processing time
Data Source
AI summary
A lane estimation apparatus including an external circumstance detection part detecting an external circumstance around a subject vehicle, and an electronic control unit including a microprocessor. The microprocessor is configured to perform estimating a current lane in which the subject vehicle travels, based on the external circumstance detected by the external circumstance detection part, and the estimating including estimating a time-based current lane every a predetermined time when the subject vehicle travels in a predetermined section, calculating a parameter indicating a frequency of the time-based current lane after traveling of the subject vehicle in the predetermined section, and estimating the current lane in which the subject vehicle has traveled in the predetermined section, based on the parameter.