Vehicle Positioning Map with Landmark-Based Feature Point Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing number of feature points in the feature point image due to objects and background within the in-vehicle camera's angle of view leads to a significant increase in data size, potentially overwhelming the memory capacity, as seen in conventional map generation devices.
Innovation Solution
A map generation apparatus that includes an in-vehicle detection unit, a microprocessor, and memory, which extracts feature points, generates maps, recognizes landmarks, and corrects the number of feature points based on their importance, thereby reducing the data size while maintaining estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of feature points is increased to improve position estimation accuracy, then the reliability of position acquisition is improved, but the data size increases and memory capacity is overwhelmed
Solution Approach 1:
The patent applies local quality by differentiating the importance of different feature points based on their spatial location and characteristics. Landmarks are identified as particularly important feature points that provide stable reference information for position estimation. By assigning different importance levels to different regions and types of feature points, the system can prioritize storage and processing of critical landmarks while reducing the number of less important feature points, thus maintaining position estimation accuracy while reducing overall data size.
2Measurement precision
If all detected feature points are stored in the map data, then the measurement precision of position estimation is improved, but the device complexity and memory requirements increase significantly
Solution Approach 1:
The patent extracts and identifies landmarks as a subset of particularly important feature points from the complete set of detected feature points. This extraction process separates critical positional reference information (landmarks) from less important feature points. By taking out only the essential landmarks for storage in the map data, the system maintains high position estimation precision while significantly reducing the data volume and memory requirements compared to storing all feature points.
3Adaptability or versatility
If the angle of view of the in-vehicle camera is widened to capture more environmental information, then the adaptability of the map generation system is improved, but the number of feature points increases and data size expands
Solution Approach 1:
The patent applies local quality by evaluating and ranking feature points based on their importance for position estimation, particularly identifying landmarks within the wide angle of view. Even though a wider angle captures more feature points, the system selectively identifies and prioritizes landmarks that provide stable and reliable positional references. This allows the system to maintain high adaptability across different environments while controlling data size by focusing on quality over quantity of feature points.
Data Source
AI summary
A map generation apparatus generating a map for use in acquiring a position of a subject vehicle. The map generation apparatus includes: an in-vehicle detection unit configured to detect a situation around a subject vehicle in traveling; a microprocessor and a memory connected to the microprocessor. The microprocessor is configured to perform: extracting one or more feature points from a detection data acquired by the in-vehicle detection unit; generating a map with the feature points extracted from the detection data in the extracting; recognizing a landmark on the map generated in the generating; determining an importance of the landmark recognized in the recognizing; and correcting a number of the feature points included in the map generated in the generating, based on the importance determined in the determining.