Vehicle Lane-Marking Cross-Check for Misrecognition Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems for automated driving fail to execute appropriate driving control based on the recognition situation of road markings, leading to potential misrecognition and unsafe driving conditions.
Innovation Solution
A vehicle control device comprising a first recognizer, a second recognizer, a comparator, and a determiner, which compare and analyze road markings from different sources to determine misrecognition and adjust driving modes accordingly, ensuring safer driving by changing tasks imposed on the occupant based on the recognition situation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single recognizer is used to identify road markings in automated driving, then the device complexity is reduced, but the reliability of road marking recognition deteriorates due to potential misrecognition
Solution Approach 1:
The patent combines multiple recognizers (first recognizer using camera images and second recognizer using map information) into a unified road marking recognition system. This merging allows cross-validation of recognition results, improving reliability while maintaining manageable system complexity through integrated control logic.
Solution Approach 2:
The system implements feedback by comparing the output of the first recognizer with the second recognizer. When discrepancies are detected, the system uses this feedback to determine misrecognition and adjust driving control accordingly, thereby improving recognition reliability without significantly increasing complexity.
2Reliability
If multiple recognizers are used to verify road markings, then the reliability of recognition is improved, but the device complexity increases
Solution Approach 1:
The recognition system is segmented into distinct functional components: a first recognizer for real-time image processing, a second recognizer for map-based verification, a comparator for result validation, and a driving controller for action execution. This segmentation improves reliability through multiple verification layers while managing complexity through modular design.
Solution Approach 2:
The comparator acts as an intermediary between the two recognizers, mediating their outputs by comparing results and determining misrecognition. This intermediary component enables reliable cross-validation without requiring direct complex interaction between all system elements, thus improving reliability while controlling overall system complexity.
3Productivity
If automated driving continues without misrecognition determination, then the productivity is maintained, but the safety deteriorates due to potential misrecognition
Solution Approach 1:
The system performs a selective misrecognition determination process that applies verification at appropriate levels. Rather than halting automated driving for every potential discrepancy, the system applies partial verification actions based on the severity and type of mismatch, maintaining productivity while improving safety through targeted safety checks.
4Reliability
If misrecognition determination process is implemented, then the safety is improved, but the loss of time increases due to additional processing
Solution Approach 1:
The system performs preliminary comparison of recognition results between the first and second recognizers before final driving control decisions are made. This preliminary action allows early detection of potential misrecognition, enabling timely safety interventions without significant time loss, as the comparison occurs in parallel with other processing tasks.
Data Source
AI summary
According to an embodiment, a vehicle control device includes a first recognizer configured to recognize a first road marking for partitioning a traveling lane of a vehicle on the basis of an output of a detection device that has detected a surrounding situation of the vehicle, a second recognizer configured to recognize a second road marking for partitioning the traveling lane with a means different from the first recognizer, a comparator configured to compare the first road marking with the second road marking, and a determiner configured to perform any one of a plurality of misrecognition determination processes including a process of determining that there is misrecognition in the first recognizer and a process of determining that there is misrecognition in one or both of the first recognizer and the second recognizer when there is a difference between the first road marking and the second road marking.