Autonomous Route Matching Across Heterogeneous SD and HD Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in integrating heterogeneous standard definition (SD) maps and high definition (HD) maps with different formats and properties, which renders incompatible map data from different producers, and existing methods fail to efficiently match and integrate these maps, leading to navigation issues such as route deviations and lane entry failures.
Innovation Solution
A system that uses a vehicle processor to generate routes by combining sensor data, wireless communication, and map information, allowing for detours based on object information from HD maps to align with global and local paths, even when map properties and formats differ.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If SD map and HD map are used from different producers with different formats and properties, then map data compatibility is poor, but using maps from the same producer ensures compatibility
Solution Approach 1:
The patent introduces a map matching unit that acts as an intermediary between SD map and HD map. This unit performs map matching by comparing map elements (roads, lanes, intersections) from both maps and identifying corresponding relationships, thereby enabling integration of heterogeneous map data from different producers without requiring them to share the same database or format.
Solution Approach 2:
The patent transforms map data from different formats into a unified representation by changing parameters such as coordinate systems, data structures, and property formats. The map matching unit adjusts and aligns parameters of SD map and HD map elements to enable compatibility between heterogeneous maps.
2Manufacturing precision
If simple matching method is used for heterogeneous maps, then processing time is short, but route accuracy deteriorates
Solution Approach 1:
The patent segments the map matching process into distinct stages: extracting map elements from SD and HD maps, comparing individual elements (roads, lanes, intersections), identifying corresponding relationships, and generating matched route information. This segmentation allows systematic processing that ensures accuracy while managing computational complexity.
Solution Approach 2:
The patent performs preliminary actions by pre-extracting and organizing map elements from both SD and HD maps before actual route matching. The map matching unit prepares reference data structures and pre-identifies key map features, which accelerates the subsequent route matching process while maintaining accuracy.
3Adaptability or versatility
If SD map and HD map use different coordinate systems and data structures, then each map maintains its own format standards, but integration between maps becomes difficult
Solution Approach 1:
The patent creates a universal map representation framework that can accommodate multiple map formats and coordinate systems. The map matching unit implements multi-functional capabilities to handle different data structures, coordinate transformations, and property formats from various map producers, enabling integrated processing of heterogeneous maps.
4Device complexity
If map data from different producers is integrated without matching, then system complexity is reduced, but navigation accuracy deteriorates
Solution Approach 1:
The map matching unit serves as an intermediary that systematically processes and matches map data from different producers. It performs structured comparison and alignment of map elements, ensuring navigation accuracy while managing system complexity through organized matching algorithms and data transformation processes.
Data Source
AI summary
An autonomous driving system may be capable of performing autonomous navigation adjustments using different maps. A user may input, on a first map, a departure point and/or a destination point. The autonomous vehicle may be configured to generate a first route using the departure point. The autonomous vehicle may initiate travel along a second route matching the first route using a different map. Based on collecting surrounding information on the second route and/or traffic information on the second route, the autonomous vehicle may be re-routed to travel along a third route.