Localisation Reference Data Using Depth Maps for Sub-Meter Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems lack the sub-meter accuracy needed for highly and fully automated driving applications, as traditional positioning methods using navigation satellites or terrestrial beacons provide only 5-10 meter accuracy, which is insufficient for precise vehicle positioning on high-definition digital maps.
Innovation Solution
Generating localisation reference data in the form of depth maps projected onto reference planes defined by navigable elements, with each pixel associated with a position and a depth channel representing the distance to environmental objects, allowing for a compressed representation of the environment that can be accurately associated with digital maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional positioning methods using navigation satellites or terrestrial beacons are used, then the system is simple and easy to operate, but the positioning accuracy is only 5-10 meters which is insufficient for automated driving
Solution Approach 1:
The patent combines multiple positioning approaches by integrating depth map-based localisation reference data with standard map data. The system merges sensor data (laser scanners, cameras) with pre-generated depth maps and digital map information to achieve sub-meter positioning accuracy, resolving the contradiction between simplicity and precision by layering computational methods over traditional positioning.
Solution Approach 2:
The patent introduces depth information as an additional dimension to traditional 2D map data. By generating and comparing depth maps (3D environmental representations) with sensor data, the system adds a vertical/depth dimension to positioning, enabling sub-meter accuracy that transcends the limitations of traditional satellite-based 2D positioning methods.
2Measurement precision
If high-definition digital maps with detailed lane representations are created, then positioning accuracy improves, but the data quantity and processing requirements increase significantly
Solution Approach 1:
The patent extracts only the essential localisation reference features needed for positioning from complete environmental data. Instead of storing and processing all sensor data, the system generates compressed depth maps that contain only the critical geometric information required for accurate positioning, reducing data quantity while maintaining sub-meter precision.
Solution Approach 2:
The patent transforms raw sensor data into a different parameter representation through depth map generation. By converting point cloud data from laser scanners into structured depth maps with specific resolutions and formats, the system changes the data parameters to achieve compact storage while preserving the positioning-critical geometric relationships.
3Measurement precision
If depth maps with high resolution are generated for localisation reference data, then positioning precision improves, but the computational complexity and processing time increase
Solution Approach 1:
The patent applies different resolution qualities to different regions of the depth maps based on their importance for positioning. Critical areas near the vehicle path use higher resolution for precision, while peripheral regions use lower resolution, optimizing the balance between positioning precision and computational complexity by allocating processing resources locally rather than uniformly.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Methods and systems for improved positioning accuracy relative to a digital map are disclosed, and which are preferably used for highly and fully automated driving applications, and which may use localisation reference data associated with a digital map. The invention further extends to methods and systems for the generation of localisation reference data associated with a digital map.