Junction Sparse Mapping With Aligned 3D Features for AV Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating due to the vast amount of data they need to process, analyze, and store, particularly when relying on traditional mapping technologies.
Innovation Solution
The system uses cameras to provide autonomous vehicle navigation features by analyzing images to detect semantic features, position descriptors, and three-dimensional feature points, and then transmits drive information to remotely located entities for creating sparse maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional mapping technology is used for autonomous vehicle navigation, then comprehensive map data can be provided, but the data processing burden and storage requirements increase significantly
Solution Approach 1:
The patent extracts only the essential navigational elements from comprehensive map data to create sparse maps. Instead of processing and storing complete traditional maps, the system identifies and extracts key features such as road boundaries, intersections, and navigable paths, thereby reducing data volume while maintaining navigation reliability
Solution Approach 2:
The patent segments the continuous road environment into discrete navigational elements and features. By dividing the complex road network into manageable segments with identified key points, the system reduces the overall data processing burden while preserving essential navigation information
2Loss of information
If comprehensive map data is stored and updated, then complete navigation information is available, but the storage requirements and data management challenges increase
Solution Approach 1:
The system extracts only the critical navigational information needed for autonomous driving from complete map data. By selecting and storing only essential elements such as road geometry, intersections, and navigable paths, the patent reduces storage volume while preventing loss of information critical for navigation
Solution Approach 2:
The patent applies partial action by storing less than the complete map data traditionally required. The sparse map contains only the portion of map information that is sufficient for navigation, eliminating the need to store and manage excessive data while maintaining navigation completeness
3Loss of information
If vast volumes of data are collected and analyzed, then comprehensive environmental information is obtained, but the processing time and computational resources increase
Solution Approach 1:
The system extracts only the essential environmental features and semantic elements needed for navigation from the vast data collected by sensors. By identifying and processing only key features such as road boundaries, traffic signals, and obstacles, the patent reduces processing time while maintaining environmental information completeness
Solution Approach 2:
The patent segments the environmental data into distinct categories and features that can be processed independently. By dividing the continuous data stream into discrete semantic elements, the system can process information more efficiently without losing comprehensive environmental understanding
Data Source
AI summary
Systems and methods for creating maps used in navigating autonomous vehicles are disclosed. In one implementation at least one processor is programmed to receive drive information from each of a plurality of vehicles that traverse different entrance-exit combinations of a road junction; for each of the entrance-exit combinations, align three-dimensional feature points in the drive information to generate a plurality of aligned three-dimensional feature point groups, one for each entrance-exit combination of the road junction; correlate one or more three-dimensional feature points in each of the plurality of aligned three-dimensional feature point groups with one or more three-dimensional feature points included in every other aligned three-dimensional feature point group from among the plurality of aligned three-dimensional feature point groups; and generate a sparse map based on the correlation, the sparse map including a target trajectory associated with each of the entrance-exit combinations.