Robotic Object Detection for Unknown Package Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems face challenges in efficiently and accurately detecting and handling unknown or unrecognized objects, particularly in de-palletizing operations, which can lead to mishandling and increased risk of injury.
Innovation Solution
A robotic system equipped with automated object detection and registration mechanisms, capable of autonomously recognizing and registering previously unknown objects by obtaining and comparing image data with registration data, and adjusting its grip and manipulation strategies accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an imaging device is employed to capture images of packages for registration, then object recognition accuracy is improved, but system complexity increases due to additional hardware and processing requirements
Solution Approach 1:
The system performs preliminary actions by capturing images of packages before handling operations and comparing them against registered images in advance. This allows the robotic system to identify physical characteristics and plan manipulation strategies beforehand, improving recognition accuracy while managing system complexity through proactive data collection and processing.
2Manufacturing precision
If manual registration of packages is performed to ensure accurate handling, then handling accuracy is improved, but productivity decreases due to manual intervention requirements
Solution Approach 1:
The robotic system performs self-service by automatically capturing images, comparing them against registered images, identifying physical characteristics, and generating manipulation strategies without human intervention. This automation maintains high handling accuracy while eliminating the need for manual registration, thereby improving productivity.
Solution Approach 2:
The system implements feedback mechanisms by continuously comparing captured package images against registered images and using the comparison results to adjust manipulation strategies. This closed-loop approach ensures accurate handling while maintaining automated operation, resolving the contradiction between handling accuracy and productivity.
3Speed
If the robotic system handles unrecognized objects without detection mechanisms, then operational speed is maintained, but safety risks increase due to potential mishandling
Solution Approach 1:
The system performs preliminary detection by capturing images and comparing them against registered images before handling operations. This preliminary identification of physical characteristics allows the robotic system to adjust manipulation strategies in advance, maintaining operational speed while ensuring safety through proactive risk assessment.
Solution Approach 2:
The detection mechanism provides feedback by continuously monitoring package characteristics and comparing them against expected parameters. When deviations are detected, the system adjusts manipulation strategies accordingly, maintaining both operational speed and safety through real-time feedback control.
Data Source
AI summary
A system and method for operating a robotic system to register unrecognized objects is disclosed. The robotic system may use first image data representative of an unrecognized object to derive an initial minimum viable region (MVR). The robotic system may analyze second image data representative of the unrecognized object to detect a condition representative of an accuracy of the initial MVR. The robotic system may register the initial MVR or an adjustment thereof based on the detected condition.