Vehicle Navigation Map Quality Index for Sensor Fusion Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles rely on high-quality sensor data, including map data, but existing methods for assessing map data quality are computationally intensive and impractical for real-time navigation, especially due to the extensive quantity of data involved.
Innovation Solution
A method and apparatus that determine a quality index for map data, weighing its reliance on other sensor data like camera and detector data, allowing for granular and localized quality assessment, and dynamically adjusting this index over time based on object type and freshness, enabling more reliable and trustworthy navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If map data quality is assessed by comparison with reference data of predefined accuracy, then measurement precision of map data quality is improved, but computational complexity and time consumption increase significantly
Solution Approach 1:
The patent introduces a quality index as an intermediary metric that indirectly represents map data quality without requiring direct comparison with reference data. This quality index is derived from sensor data characteristics and serves as a computationally efficient proxy for quality assessment, resolving the contradiction between measurement precision and computational complexity
Solution Approach 2:
The patent creates a simplified representation (quality index) that copies the essential quality information of map data without requiring the full reference data comparison process. This copied quality metric enables rapid assessment while maintaining sufficient accuracy for autonomous navigation decisions
2Reliability
If map data quality assessment is performed in real-time for autonomous navigation, then navigation reliability is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary quality assessment by continuously monitoring sensor data characteristics and updating the quality index in advance of navigation decisions. This preliminary action enables rapid quality evaluation at the moment of navigation planning, improving reliability without adding processing delays
Solution Approach 2:
The patent maintains continuous quality assessment through ongoing sensor data processing and quality index updates. This continuous monitoring ensures navigation reliability is consistently evaluated without periodic interruptions or batch processing delays, enabling real-time adaptive navigation
3Ease of operation
If generalized high level quality statements are used for map data, then ease of operation is improved, but navigation reliability for autonomous vehicles deteriorates
Solution Approach 1:
The patent transitions from generalized quality statements to localized quality assessment by computing quality indices for specific geographic locations and map features. This local quality approach provides detailed, location-specific reliability information that enhances autonomous navigation decisions while maintaining operational simplicity through automated computation
Solution Approach 2:
The patent changes the quality representation from qualitative general statements to quantitative localized parameters (quality indices). This parameter transformation enables precise reliability assessment for different map regions and features, improving navigation reliability while keeping the system easy to operate through automated index calculation
Data Source
AI summary
An apparatus, method and computer program product are provided to facilitate the navigation of a vehicle, such as an autonomous vehicle, utilizing map data in which the quality associated with the map data is provided in a more computationally efficient manner. In the context of a method a plurality of different types of sensor data are received including map data, camera data and detector data. The method determines a quality index associated with the map data and weights the reliance upon the map data relative to other types of sensor data based upon the quality index associated with the map data. The method further includes determining navigation information for the vehicle based at least partly upon the weighting of the map data relative to other types of sensor data.