Driving Support Apparatus Lane Boundary Error Threshold
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving support systems face accuracy degradation when distinguishing between marking locations and lane boundaries, especially on branched roads, leading to suboptimal notifications and vehicle controls.
Innovation Solution
A driving support apparatus equipped with an image capturing unit, lane recognition unit, travelling road calculation unit, control process execution unit, and error calculation unit, which estimates lane locations and shapes using recognized boundary lines and adjusts vehicle control based on error thresholds to improve accuracy and notification reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the recognized marking location is regarded as a current location of the lane boundary and future location is estimated using the current location, then the system can provide basic lane keeping control, but estimation accuracy is degraded on branched roads
Solution Approach 1:
The patent segments the lane boundary estimation into two distinct components: actual boundary line detection (using image recognition of markings) and estimated boundary line calculation (using traveling road parameters). This segmentation allows the system to differentiate between directly observed data and inferred data, enabling more accurate handling of branched road scenarios where markings may not reflect the true lane boundary.
Solution Approach 2:
The patent introduces traveling road parameters as an intermediary element between the recognized marking location and the estimated lane boundary. These parameters act as a mediator that incorporates additional contextual information about the road geometry and vehicle trajectory, improving estimation accuracy without requiring direct observation of the entire lane boundary.
2Ease of operation
If the system performs vehicle control based on estimated lane boundary, then driving support functionality is provided, but control accuracy deteriorates when markings and lane boundaries appear similar on branched roads
Solution Approach 1:
The patent implements a feedback mechanism where the system continuously compares the actual boundary line (from image recognition) with the estimated boundary line (from traveling road parameters). This feedback loop allows the system to detect discrepancies and adjust control actions accordingly, maintaining high accuracy even when markings and lane boundaries appear similar on branched roads.
Solution Approach 2:
The patent makes the lane boundary estimation dynamic by continuously updating traveling road parameters based on current vehicle position, speed, and steering angle. This dynamic approach allows the system to adapt to changing road conditions and maintain accurate vehicle control even in complex branched road scenarios where static marking-based estimation would fail.
Data Source
AI summary
A driving support apparatus is provided to improve accuracy of the driving support such as a notification to the driver and the vehicle control. In the apparatus, a common parameter indicating a location of a lane boundary is calculated for the vehicle control, based on an actual boundary line where a location of the lane boundary line existing at both sides of the vehicle is actually measured. A common error is defined as an error between the location of an estimated boundary indicated by the common parameter and a location of the actual boundary line. In the case where the common error is larger than or equal to a common threshold in which an amount of the common error is set in advance, the apparatus determines that the travelling road where the vehicle travels is branched and stops the vehicle control using the common parameter.