Map Information System Road Marking Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving support control systems face inaccuracies in detecting specific road markings like stop lines and pedestrian crossings due to reliance on brightness distribution, leading to erroneous detections and low reliability calculations.
Innovation Solution
A map information system that uses a database management device to detect road marking candidates based on driving environment information, recognizes vehicle behavior around these candidates, and determines an evaluation value to confirm the presence of specific road markings, thereby increasing the accuracy of road marking map information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If brightness distribution is used to detect road markings, then detection process is simple, but detection accuracy is low and erroneous detections occur
Solution Approach 1:
The system uses vehicle behavior feedback (deceleration, stopping patterns) to verify and correct road marking detections. When a road marking is detected, the system monitors subsequent vehicle behavior to confirm whether the detection is correct, creating a closed-loop verification process that improves accuracy without significantly increasing initial detection complexity
Solution Approach 2:
Vehicle behavior data serves as an intermediary to bridge the gap between simple brightness-based detection and accurate road marking identification. The detection system first identifies candidates using simple brightness analysis, then uses vehicle behavior patterns as an intermediary verification layer to filter out erroneous detections and confirm valid ones
2Measurement precision
If vehicle behavior recognition is added to verify road markings, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The system applies vehicle behavior recognition selectively rather than universally. It performs full verification only for road marking candidates that require confirmation, using partial action (behavior analysis) only where needed to resolve ambiguity, thereby improving accuracy while minimizing the increase in system complexity
Solution Approach 2:
The system performs preliminary brightness-based detection to identify road marking candidates before applying the more complex vehicle behavior verification. This preliminary action filters out obvious cases and focuses the complex verification process only on ambiguous candidates, reducing the overall computational burden
Data Source
AI summary
A map information system includes a database management device configured to manage a map database used for vehicle driving support control. The map database includes road marking map information that indicates a position of a specific road marking including at least one of a stop line and a pedestrian crossing. The database management device is further configured to: detect a road marking candidate being a candidate for the specific road marking around a vehicle; recognize vehicle behavior of at least one of the vehicle and another vehicle in a period when passing the road marking candidate; determine, based on the vehicle behavior, an evaluation value that indicates certainty of the road marking candidate being the specific road marking; and register the road marking candidate having the evaluation value equal to or higher than a threshold, as the specific road marking, in the road marking map information.