Position-Posture Estimation via Frame Selection and Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional position-posture estimation methods in AR and AGV systems face accuracy issues due to accumulated errors in relative position-posture calculations and variability in absolute position-posture accuracy based on image patterns or shape characteristics, leading to reduced precision, especially with featureless or iterative patterns.
Innovation Solution
A position-posture estimation device and method that integrate data from multiple image frames captured from different viewpoints, selecting frames based on relative movement and variance, and calculating absolute position-posture by combining relative and absolute postures, thereby enhancing accuracy through frame selection and integration processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only one image frame is used for absolute position-posture calculation, then the computation is simple and fast, but the accuracy varies greatly depending on the subject patterns
Solution Approach 1:
The patent segments the image sequence into multiple candidate frames and evaluates each frame's suitability for absolute position-posture calculation independently. By dividing the problem into frame-level assessments and selective integration, the system achieves higher accuracy without overwhelming computational complexity.
Solution Approach 2:
The patent changes the parameter of frame selection by evaluating multiple frames based on their characteristic values (such as pattern distinctiveness) and selecting only those that meet predetermined criteria. This parameter-based filtering approach maintains computational efficiency while significantly improving accuracy by excluding frames with problematic patterns.
2Measurement precision
If multiple image frames are processed for position-posture calculation, then the accuracy is improved, but the computation time and processing load increase
Solution Approach 1:
The patent performs preliminary evaluation of each frame's characteristic value before full position-posture calculation. Frames that fail the preliminary criterion are discarded immediately, avoiding unnecessary computation. This preliminary filtering action maintains real-time processing capability while improving overall accuracy.
Solution Approach 2:
The patent processes only a subset of frames (those meeting the predetermined criterion) rather than all frames in the sequence. This partial processing approach reduces computational load while still achieving improved accuracy through the integration of multiple qualified frames.
3Speed
If relative position-posture calculation is executed at constant short cycle, then the responsiveness is high, but accumulated error becomes significant in long-distance movement
Solution Approach 1:
The patent implements a feedback mechanism where absolute position-posture calculations (based on image frames) are periodically integrated into the relative position-posture estimation. This feedback corrects accumulated errors from the high-speed relative calculation, maintaining both responsiveness and long-term accuracy.
Solution Approach 2:
The patent merges two calculation approaches: high-speed relative position-posture calculation (for responsiveness) and more accurate absolute position-posture calculation from selected image frames (for accuracy). By combining these methods, the system achieves both fast response and reduced accumulated error over long distances.
Data Source
AI summary
A position-posture estimation device includes processing circuitry to read in data of a three-dimensional map from a database; to execute a process of selecting a frame to be used for calculation of a position-posture from a plurality of image frames captured from different viewpoints; to execute a process of acquiring a plurality of relative position-postures regarding a plurality of frames selected by the frame selection unit; to execute a process of acquiring a plurality of absolute position-postures regarding the plurality of selected frames; and to acquire a final absolute position-posture by integrating the acquired relative position-postures and the acquired absolute position-postures.