ADAS Lane Recognition Failure Detection via Multi-Camera Cross-Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced driver assistance systems (ADAS) face increased accident risks due to lane recognition failures, which can lead to hardware and software malfunctions, and misrecognition issues, compromising safety and autonomous driving capabilities.
Innovation Solution
An advanced driver assistance device equipped with multiple image acquisition units and lane recognizers that compare lane information across different images to detect failures, allowing for real-time correction and output of failure information, thereby reducing the risk of accidents and maintaining autonomous driving control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single lane recognizer and image acquisition unit are used, then the device complexity is low, but the reliability of lane recognition decreases due to potential failures
Solution Approach 1:
The patent applies local quality by having different image acquisition units capture images from different positions (front, rear, left, right) of the vehicle. Each unit has a specific local function for capturing road images from its designated position, which when combined, provides comprehensive lane recognition coverage and improves overall system reliability
Solution Approach 2:
The patent merges multiple image acquisition units and lane recognizers into a unified system controlled by a single controller. The controller integrates lane information from all units, compares their outputs, and determines failures by analyzing inconsistencies, thereby improving reliability through redundancy while managing complexity through centralized control
2Reliability
If multiple image acquisition units and lane recognizers are used, then the reliability of lane recognition improves, but the device complexity increases
Solution Approach 1:
The controller performs multiple functions: it processes lane information from all image acquisition units, compares their outputs, determines failures, and controls the output unit. This multi-functionality reduces the need for separate dedicated components for each task, thereby managing system complexity while maintaining high reliability through redundant recognition
Solution Approach 2:
The system implements feedback by having the controller continuously compare lane information from multiple recognizers and use this comparison to determine failures. The feedback loop allows the system to self-diagnose and maintain reliability by identifying and compensating for failures in individual components
3Reliability
If lane recognition information is not verified, then the processing speed is fast, but the safety decreases due to undetected failures
Solution Approach 1:
The system performs preliminary verification by having multiple lane recognizers process images simultaneously and compare their outputs before final lane determination. This preliminary cross-verification detects failures early in the processing chain, ensuring safety without requiring sequential processing that would increase time loss
Data Source
AI summary
An advanced driver assistance apparatus, a vehicle having the same, and a method of controlling the vehicle are provided. The vehicle may acquire a first image of a road by a first image acquisition unit while traveling, acquire a second image of the road by a second image acquisition unit while traveling, recognize a first lane in the first image by a first lane recognizer, recognize a second lane in the second image by a second lane recognizer, determine whether at least one of the first lane recognizer or the first image acquisition unit has a failure by comparing the first lane with the second lane, and output failure information when at least one of the first lane recognizer or the first image acquisition unit is determined to have the failure.