Map Data Adaptation Using Object Feature Reference Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for creating map data struggle when real-world road conditions, such as construction sites or shifted lanes, do not match the stored map data, leading to inaccuracies and unreliable vehicle guidance in driver assistance systems.
Innovation Solution
A method that receives vehicle trajectories and object features, checks for existing map data, and adapts the map data using a weighting factor to correct inaccuracies, averaging data from multiple vehicles for increased accuracy, especially in areas with shifted lanes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If map data are created from vehicle trajectories alone, then map data can be generated for areas without existing map data, but the accuracy deteriorates when satellite navigation is inaccurate causing absolute positioning errors
Solution Approach 1:
The patent introduces object features (such as road signs, buildings, or other stationary objects) as an intermediary reference system. Instead of relying solely on satellite navigation coordinates, the system uses these objects as mediators to establish accurate spatial relationships. The object features serve as a common reference frame that both the vehicle trajectory and map data can be aligned to, thereby eliminating the absolute positioning errors caused by satellite navigation inaccuracy.
Solution Approach 2:
The system implements feedback by continuously comparing detected object features with stored map data and using this comparison to correct trajectory deviations. When a vehicle passes by known objects, the system measures the actual position relative to these objects and uses this feedback information to adjust and correct the map data, ensuring long-term accuracy even as roads and lanes may shift.
2Adaptability or versatility
If map data are updated frequently to reflect real-world changes, then the map data remain current with construction sites and shifted lanes, but the system complexity increases
Solution Approach 1:
The system performs self-updating by automatically detecting deviations between actual vehicle trajectories (corrected by object features) and stored map data. When deviations exceed predefined thresholds, the system autonomously updates the map data without requiring manual intervention or complex external verification systems. This self-service mechanism maintains map currentness while keeping system complexity manageable.
Solution Approach 2:
Instead of continuously updating all map data, the system applies updates only where and when deviations are detected beyond threshold values. This partial action approach updates only the necessary portions of map data (specific lanes or road sections) rather than performing full system updates, thereby maintaining adaptability while reducing overall system complexity and computational burden.
3Measurement precision
If vehicle trajectories from multiple vehicles are averaged, then the statistical accuracy improves, but the processing time increases
Solution Approach 1:
The system performs preliminary correction of individual vehicle trajectories using object features before averaging. By pre-aligning each trajectory to the object feature reference frame, the system eliminates the need for complex post-processing and reduces the computational burden during averaging. This preliminary action ensures that trajectories are already in a consistent reference system, enabling faster and more accurate statistical processing.
Data Source
AI summary
A method for creating map data having lane-specific resolution. Mapping data are initially received, the mapping data being transmitted by a vehicle, and including a vehicle trajectory and at least one object feature. Once the mapping data has been received, it is checked whether map data for local surroundings of the received mapping data are present. In the event the check indicates that no map data are present, map data are created from the mapping data and stored in a memory. In the event the check indicates that map data are already present, the map data are compared with the mapping data. If this comparison reveals that the mapping data differ from the map data, the map data are adapted, the adaptation taking place on the basis of a weighting factor. The adapted map data are subsequently stored in the memory.