Robot Repositioning Using Similar Scene Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot repositioning technologies are prone to errors due to the reliance on map frames with maximum matching degrees, leading to inaccurate positioning.
Innovation Solution
A method involving the acquisition of environmental image data, extraction of global description information, and determination of K map frames with maximum similarity in a visual map, followed by repositioning based on a map frame with maximum similarity when no symmetrical or multiple similar scenes are present, using neural networks for feature extraction and PnP algorithm for positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If robot repositioning relies on map frames with maximum matching degrees, then the repositioning process is simple and fast, but the accuracy deteriorates due to errors in positioning
Solution Approach 1:
The patent segments the repositioning process into multiple validation stages: initial map frame selection based on global description information, followed by symmetrical scene detection, and multiple similar scene verification. This segmentation allows the system to maintain simplicity in each individual stage while achieving high overall accuracy through cumulative validation.
Solution Approach 2:
The patent implements feedback mechanisms at multiple levels: the system continuously validates the selected map frame against symmetrical scene criteria and multiple similar scene criteria, using the results to confirm or reject the repositioning result. This feedback loop ensures accuracy without requiring overly complex processing at any single stage.
2Measurement precision
If the system checks for symmetrical scenes and multiple similar scenes, then the repositioning accuracy is improved, but the computation time increases
Solution Approach 1:
The patent performs preliminary actions by extracting global description information from environmental image data before the actual repositioning calculation. This preliminary feature extraction and map frame pre-filtering based on global description information reduces the computation time required during the subsequent symmetrical scene and similar scene validation stages.
Solution Approach 2:
The patent applies different quality thresholds and validation criteria at different stages: loose criteria for initial map frame selection, moderate criteria for symmetrical scene detection, and strict criteria for multiple similar scene verification. This local quality differentiation optimizes computation time by applying heavy validation only where necessary.
3Measurement precision
If the system uses neural networks for feature extraction, then the repositioning accuracy is improved, but the device complexity and computational resources increase
Solution Approach 1:
The patent extracts only the essential global description information from the environmental image data using neural networks, rather than processing the complete image data. This extraction approach maintains high repositioning accuracy while reducing the computational burden and system complexity by working with condensed feature representations rather than raw images throughout the entire pipeline.
Data Source
AI summary
In one aspect, a repositioning method includes: obtaining environment image data of a target object to be positioned at a current position; extracting global description information from the environment image data; determining K map frames having the highest similarity with the environment image data from a visual map according to the global description information; when it is determined that the current position of the target object does not have a symmetric scene or a plurality of similar scenes in the visual map according to the K map frames having the highest similarity, if it is determined that a map frame having the highest similarity is a target similar scene of the current position, repositioning the target object according to the map frame having the highest similarity.