Robot Depth Filtering for High-Position Object Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques struggle to accurately identify and handle objects at higher positions within a pile, especially when objects are randomly stacked, due to limitations in depth sensing and object recognition.
Innovation Solution
An information processing device and robot control system that utilize depth information filtering based on object size to generate high-position information, enabling the identification of high portions of objects and determining the optimal holding posture for a robot to handle these objects effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If depth information is obtained without filtering, then all objects in the measurement space are captured, but it becomes difficult to accurately identify high-position objects among randomly stacked objects
Solution Approach 1:
The patent extracts only the relevant portion of depth information by applying filtering processing to obtain high-position information. The filtering process selectively extracts depth data corresponding to objects at higher positions, separating them from the complete depth information of all objects. This extraction approach improves identification accuracy while managing processing complexity through targeted data selection.
Solution Approach 2:
The patent segments the measurement space into multiple depth layers by applying filtering to depth information. The filtering process divides objects into different height groups, allowing the system to specifically identify and process high-position objects separately from other objects. This segmentation enables precise targeting of high-position objects without processing the entire dataset.
2Measurement precision
If filtering is applied to depth information based on object size, then high-position objects can be accurately identified, but processing time increases
Solution Approach 1:
The patent applies filtering processing selectively to depth information rather than processing all depth data equally. By focusing the filtering operation specifically on extracting high-position information, the system achieves accurate identification while minimizing unnecessary processing. This partial action approach processes only the essential data needed for high-position object identification.
3Ease of operation
If the robot attempts to handle objects without accurate high-position information, then the system is simpler, but the robot cannot determine appropriate holding postures for randomly stacked objects
Solution Approach 1:
The patent introduces high-position information as an intermediary between the depth sensor and the robot control system. This intermediate processed information serves as a bridge that translates raw depth data into actionable insights for the robot. The high-position information acts as a mediator that enables the robot to determine appropriate holding postures without requiring the robot itself to perform complex depth analysis.
Data Source
AI summary
An information processing device includes an obtainer and a generator. The obtainer obtains depth information indicating a depth in a measurement space. The measurement space includes a plurality of objects. The generator generates high-position information indicating a high portion of the plurality of objects at a higher position than another portion of the plurality of objects in the measurement space by performing, on the depth information, filtering corresponding to an object size of an object of the plurality of objects.