Sparse Lane Map Navigation With Crowdsourced Road Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating with vast volumes of data, including image data, GPS data, and sensor data, which can limit navigation efficiency and require excessive storage and update of traditional maps.
Innovation Solution
The use of cameras to construct and navigate with a sparse map, supplemented by GPS data and sensor data, allowing for efficient data storage and adaptive navigation using a crowdsourced sparse map that includes lane measurements and road surface features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional mapping technology is used for autonomous vehicle navigation, then navigation accuracy is maintained, but data storage requirements and map update complexity increase excessively
Solution Approach 1:
The patent extracts only the essential navigation-critical information from traditional dense maps, including 3D building models, road geometry, lane markings, and points of interest, while discarding redundant visual and descriptive data. This extraction creates a sparse map that maintains navigation accuracy while dramatically reducing storage requirements and update complexity.
Solution Approach 2:
The patent segments the map data into hierarchical levels, separating critical navigation elements (roads, buildings, lanes) from secondary information. This segmentation allows the system to store and process only the essential structural data needed for navigation, reducing overall data volume while preserving navigational functionality.
2Loss of information
If traditional mapping technology is used, then comprehensive navigation information is available, but map update complexity and computational burden increase
Solution Approach 1:
The patent extracts only the essential navigation-critical information from traditional dense maps, including 3D building models, road geometry, lane markings, and points of interest, while discarding redundant visual and descriptive data. This extraction creates a sparse map that maintains navigation accuracy while dramatically reducing storage requirements and update complexity.
Solution Approach 2:
The patent implements a dynamic map update mechanism that selectively updates only the affected segments of the sparse map when new data is collected, rather than updating the entire map. This dynamic approach reduces computational burden and update complexity while maintaining information completeness.
3Loss of information
If dense map data is used for navigation, then detailed environmental information is available, but processing time and computational resources increase
Solution Approach 1:
The patent extracts only the essential navigation-critical information from traditional dense maps, including 3D building models, road geometry, lane markings, and points of interest, while discarding redundant visual and descriptive data. This extraction creates a sparse map that maintains navigation accuracy while dramatically reducing storage requirements and update complexity.
Solution Approach 2:
The patent segments the map data into hierarchical levels, separating critical navigation elements (roads, buildings, lanes) from secondary information. This segmentation allows the system to process only essential data structures during navigation, reducing computational time while preserving necessary environmental information.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Method, system and computer-readable medium for autonomously navigating a vehicle along a road segment, comprising: receiving a sparse map model, wherein the sparse map model includes at least one line representation of a road surface feature extending along the road segment, each line representation representing a path along the road segment substantially corresponding with the road surface feature; receiving from a camera, at least one image representative of an environment of the vehicle; analyzing the sparse map model and the at least one image received from the camera to determine a current position of the vehicle relative to a longitudinal position along the at least one line representation of the road surface feature extending along the road segment; and determining an autonomous navigational response for the vehicle based on the analysis of the sparse map model and the at least one image received from the camera.