Road Object Data Validation Using Ground Truth States
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems face challenges in accurately detecting road objects in complex road geometries, leading to unreliable road object data, which can result in false positives or negatives when validating road object data.
Innovation Solution
The system uses ground truth data collected by manually driven vehicles to validate road object data, determining its state as either partial or complete to perform specific validation operations, ensuring accurate association of road objects with their respective links and providing reliable navigation functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ground truth data is collected by manually driven vehicles, then the reliability of validation data is improved, but the ground truth data may be incomplete due to alternative links not being covered
Solution Approach 1:
The system performs partial validation when ground truth data is incomplete by validating only against the portions of road object data that correspond to covered links, rather than requiring complete coverage of all alternative links. This allows the validation process to proceed with available data while acknowledging its limitations.
Solution Approach 2:
The system changes the state parameter of ground truth data from incomplete to complete by collecting additional data from multiple ground truth vehicles traversing different routes, including alternative links. This transforms the validation process from partial to complete when sufficient data is available.
2Loss of information
If ground truth vehicles cover all alternative links and main link, then complete ground truth data is obtained, but the complexity and cost of data collection increases
Solution Approach 1:
The validation process is segmented into two distinct operations: partial validation for incomplete ground truth data and complete validation for complete ground truth data. This segmentation allows the system to adapt its validation strategy based on data availability, reducing the need for always-collecting-complete-data approaches.
Solution Approach 2:
The validation system is designed to perform multiple functions: it can execute partial validation when data is incomplete and complete validation when data is sufficient. This multi-functionality allows a single system to handle varying data completeness scenarios without requiring separate specialized systems.
3Productivity
If partial ground truth data is used for validation, then the validation process can proceed with available data, but false positives or negatives may occur
Solution Approach 1:
The validation system dynamically adjusts its operation mode based on the completeness of ground truth data. When data is incomplete, it executes partial validation with appropriate limitations; when data becomes complete, it switches to complete validation. This dynamic adaptation optimizes both efficiency and reliability based on current data conditions.
Solution Approach 2:
The system incorporates feedback mechanisms that monitor the completeness of ground truth data and adjust the validation strategy accordingly. The feedback loop determines whether to perform partial or complete validation, ensuring that the validation process matches the quality and completeness of available data.
Data Source
AI summary
A system for validating the road object data is provided. The system, for example, receives the road object data for a geographic region and determines, from a map database, a state of ground truth data associated with the geographic region. The state of the ground truth data comprises at least one of partial ground truth data and complete ground truth data. Further, the road object data is validated based on the determined state of the ground truth data. The validation of the road object data includes performing a first validation operation when the state of the ground truth data is the partial ground truth data and performing a second validation operation when the state of the ground truth data is the complete ground truth data.