Autonomous Vehicle Position Estimation Using Prioritized Map Image Groups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The image matching process in autonomous vehicles becomes lengthy when using all map image datasets, leading to reduced processing time and lower position estimation accuracy due to changes in the image-capturing environment.
Innovation Solution
Divide map datasets into groups based on the imaging subject's state, rearrange their order to prioritize groups with similar image data, and execute the image matching process using only a predetermined number of datasets from the head of the new order.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the image matching process uses all map image datasets stored in the storage device, then the position estimation accuracy is improved, but the processing time becomes excessively long
Solution Approach 1:
The patent segments the map image datasets into multiple groups based on image-capturing environment characteristics (e.g., daytime/nighttime, weather conditions, road types). Instead of processing all datasets uniformly, the system divides them into manageable segments and selectively processes only the relevant groups, thereby reducing processing time while maintaining position estimation accuracy through targeted matching.
2Loss of time
If the number of map image datasets used is reduced to shorten processing time, then the processing time is improved, but the position estimation accuracy deteriorates
Solution Approach 1:
The patent applies local quality by assigning different processing priorities to different groups of map image datasets based on their relevance to the current image-capturing environment. Highly relevant groups (matching current conditions) are processed with higher priority and greater detail, while less relevant groups are processed with lower priority or skipped entirely. This ensures that processing resources are concentrated on the datasets most likely to yield accurate position estimates.
3Productivity
If the arrangement order of map datasets is changed to prioritize datasets closer to the estimated position, then the processing efficiency is improved, but the position estimation accuracy may deteriorate due to environmental differences
Solution Approach 1:
The patent performs preliminary classification of map image datasets into groups based on image-capturing environment characteristics before the actual image matching process. This preliminary action ensures that when datasets are arranged and selected for processing, the most environmentally relevant groups are already positioned and ready for priority processing, eliminating the need to reconsider environmental compatibility during the matching phase and ensuring both efficiency and accuracy.
Data Source
AI summary
An autonomous traveling vehicle performs data alignment processing, priority order setting processing, identification processing, image matching processing, and own-position estimation processing. The data alignment processing rearranges a plurality of map data into a prescribed order. The identification processing identifies, as an identified group, a group that includes map image data that is most similar to image data when a prescribed condition is satisfied after the data alignment processing. The priority order setting processing further rearranges the plurality of map data rearranged by the data alignment processing, so that the map data belonging to the identified group is used preferentially in the image matching processing.