Digital Road Map Object Removal Using Multi-Vehicle Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Maintaining the accuracy and currency of digital road maps used in advanced driver assistance systems and autonomous vehicles is challenging due to the need for continuous updates with current road and traffic sign information.
Innovation Solution
An object updating system that includes a backend configured to receive environmental data from vehicles, recognize objects in the data, and remove objects from the digital road map if they cannot be recognized, using probability values and boundary conditions to determine object existence and visibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If digital road maps are continuously updated with new objects from multiple vehicles, then the currency and accuracy of the road map improves, but the complexity of data processing and verification increases
Solution Approach 1:
The system performs preliminary actions by having vehicles capture and transmit environmental data continuously, so that when an object needs to be verified or removed, the data is already available in the backend system for immediate processing without requiring new data collection campaigns
Solution Approach 2:
The backend system acts as an intermediary that receives environmental data from multiple vehicles, processes the information centrally, and manages the digital road map updates. This intermediary approach simplifies individual vehicle systems while maintaining high overall system reliability through centralized coordination
2Measurement precision
If objects are removed from the digital road map when not recognized, then the accuracy of the road map improves, but the risk of removing valid objects due to temporary detection failures increases
Solution Approach 1:
The system checks boundary conditions preliminarily before removing objects, including verifying whether the vehicle was at appropriate distances and angles to detect the object, and whether environmental conditions (weather, time, season) were favorable for detection. This preliminary verification reduces the risk of falsely removing valid objects
Solution Approach 2:
The system uses feedback from multiple vehicles and multiple observation opportunities to verify object existence before removal. If an object is not detected by one vehicle, the system checks whether other vehicles have detected it or whether the non-detection can be explained by boundary conditions, providing a feedback mechanism that prevents premature removal of valid objects
3Reliability
If the system considers multiple boundary conditions for object recognition, then the reliability of object detection improves, but the computational requirements and processing time increase
Solution Approach 1:
The system performs preliminary checks of boundary conditions (distance, angle, weather, time) before conducting full object recognition analysis. This preliminary filtering reduces the computational burden by quickly eliminating cases where detection failure is expected due to unfavorable conditions, allowing the system to focus processing resources only on ambiguous cases that require detailed analysis
Data Source
AI summary
Various embodiments include a backend for an object updating system for removing objects present in a digital road map comprising: a memory storing a set of instructions; and a processor in communication with the memory. The digital road map is stored in the memory. The set of instructions, when loaded and executed by the processor, cause the processor to: receive a path and environmental data captured along this path from a vehicle; recognize objects in the environmental data; and remove an object from the digital road map if the object is not recognized in the environmental data.