Object State Identification Using Virtual Plane Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object state identification methods face a trade-off between speed and accuracy, making it difficult to achieve both high speed and high accuracy in identifying the position and posture of objects, especially when using three-dimensional information from images.
Innovation Solution
The method involves obtaining actual measured three-dimensional values at multiple points on a flat portion of an object with an opening, extracting corresponding point values to determine a virtual plane, and identifying the object state based on stable points within this plane, using a sensor and a processor to quickly and accurately determine the object's position and posture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional object state identification methods using three-dimensional information from images are employed, then measurement precision can be improved, but calculation load increases and identification speed decreases
Solution Approach 1:
The patent extracts only the necessary information (opening edges and flat portion geometry) from the complete three-dimensional object data. By focusing on specific geometric features rather than processing all point cloud data, the system achieves accurate object state identification while significantly reducing calculation load and improving identification speed.
Solution Approach 2:
The patent segments the object identification process into distinct stages: first identifying opening edges in the image, then determining the flat portion plane equation, and finally calculating object state parameters. This segmentation allows each stage to process only relevant data, improving overall efficiency while maintaining precision.
2Measurement precision
If feature points are extracted from three-dimensional information to identify object state, then measurement precision improves, but calculation complexity increases
Solution Approach 1:
Instead of extracting and processing numerous feature points from three-dimensional data, the patent extracts only the opening edges and uses them to determine the flat portion plane. This extraction approach maintains identification accuracy while dramatically simplifying the calculation process by reducing the number of data points that need processing.
Solution Approach 2:
The patent transforms the complex three-dimensional point cloud data into simplified parameters: the plane equation of the flat portion and the positions of opening edges. By representing the object geometry through these fewer parameters, the system achieves the same identification accuracy with much lower computational complexity.
Data Source
AI summary
A carrier for carrying an object includes a fork that is vertically movable to carry the object and that is able to be inserted into an opening of a flat portion of the object, a sensor that obtains actual measured values at a plurality of points on the flat portion of the object, and an object state identification device that identifies, based on the actual measured values, an object state including at least one of a position and a posture of the object with respect to the sensor.