Occupancy Grid Map Updating with Influence-Based Scan Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The production and updating of occupancy grid maps in SLAM systems consume increasing memory and processing resources as scan data accumulates, leading to potential loss of effective information and reduced map accuracy due to the overwriting of old scan data.
Innovation Solution
A map generation system that calculates an influence ratio for detection information and sorts it based on this ratio to prioritize and retain only high-influence data, reducing the number of data points needed for processing and maintaining map accuracy while minimizing memory and processing loads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If scan data is accumulated to improve map accuracy, then map accuracy is improved, but memory consumption and arithmetic processing load increase
Solution Approach 1:
The patent extracts only the essential and effective information from scan data by calculating an influence ratio for each piece of detection information. The map production device selectively uses only those pieces of detection information with high influence ratios to produce the occupancy grid map, thereby reducing memory consumption while maintaining map accuracy.
Solution Approach 2:
The patent introduces a new parameter - the influence ratio - to evaluate the effectiveness of each piece of detection information. By calculating and sorting detection information based on this parameter, the system can dynamically determine which data to retain and which to discard, optimizing the balance between map accuracy and memory usage.
2Measurement precision
If scan data is accumulated to improve map accuracy, then map accuracy is improved, but arithmetic processing load increases
Solution Approach 1:
The patent extracts only the essential and effective information from scan data by calculating an influence ratio for each piece of detection information. The map production device selectively uses only those pieces of detection information with high influence ratios to produce the occupancy grid map, thereby reducing arithmetic processing load while maintaining map accuracy.
Solution Approach 2:
Instead of processing all scan data, the patent applies partial action by selectively processing only the most influential pieces of detection information. This approach achieves sufficient map accuracy without the excessive computational burden of processing every available data point.
3Quantity of substance
If old scan data is deleted to reduce memory usage, then memory consumption is reduced, but effective information is lost and map accuracy degrades
Solution Approach 1:
The patent introduces an influence ratio parameter to evaluate each piece of detection information regardless of its age. This allows the system to retain old scan data that still has high influence (effective information) while discarding old data with low influence, thereby reducing memory consumption without degrading map accuracy.
Solution Approach 2:
The patent applies different retention policies to different pieces of old scan data based on their individual influence ratios. Rather than uniformly deleting all old data, the system selectively retains those pieces that contribute most to map accuracy, achieving local optimization of memory usage.
4Measurement precision
If all detection information is used to produce occupancy grid map, then map accuracy is improved, but the number of data points for processing increases
Solution Approach 1:
The patent extracts only the most influential pieces of detection information by calculating influence ratios and sorting the data. The map production device uses only the top-ranked detection information with high influence ratios, significantly reducing the number of data points that need to be processed while maintaining map accuracy.
Solution Approach 2:
The patent applies partial action by using only a subset of detection information - specifically, those pieces with the highest influence ratios. This approach achieves sufficient map accuracy without the complexity of processing all available data points.
Data Source
AI summary
Provided are a map generation system and a mobile object configured so that the map accuracy of an occupancy grid map can be ensured while a memory capacity and an arithmetic processing load can be reduced. The map generation system includes a storage (27) configured to store, as detection information, a detection value obtained for each detection unit by a detection device (4) in chronological order, a map producer (23) configured to produce the occupancy grid map based on the detection information stored in the storage (27) and cause the storage (27) to store the occupancy grid map, and an influence ratio calculator (24) configured to calculate an influence ratio for the occupancy grid map for each piece of detection information stored in the storage (27). The map producer (23) sorts the detection information based on the influence ratio calculated by the influence ratio calculator (24) to produce the occupancy grid map.