Robotic Package Scan and Registration for Unknown Object Handling

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current robotic systems face challenges in efficiently handling and registering unknown or unrecognized objects during de-palletization, leading to potential mishandling and increased costs due to manual intervention and errors in identifying physical characteristics.

Innovation Solution

A robotic system capable of autonomously scanning and registering previously unknown objects by identifying exposed edges and corners, deriving minimum viable regions for gripping, and obtaining additional information such as heights and identifying marks, allowing for accurate manipulation and registration without operator assistance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated robotic de-palletization is implemented, then productivity increases and manual intervention is reduced, but the system may fail to correctly identify physical characteristics of unknown objects, leading to mishandling and stoppages

Engineering Contradiction:

Improvede-palletization throughputVSAvoidobject identification accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The system performs preliminary scanning and registration of objects before they are manipulated by the robotic arm. Imaging devices capture images of objects on the pallet, and the system processes these images to identify physical characteristics such as dimensions, weight, and center of mass before the robotic de-palletization process begins, ensuring accurate handling from the start

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system continuously compares captured images of objects with registered images stored in a database. When discrepancies are detected or objects are unrecognized, the system can pause and perform additional scanning or manual registration, using feedback from the image comparison process to ensure accurate identification before proceeding with manipulation

Inventive Principle:

Principle #23Feedback

2Measurement precision

If the system pauses for manual registration of unrecognized objects, then identification accuracy improves, but productivity decreases due to stoppages

Engineering Contradiction:

Improvephysical characteristics identification accuracyVSAvoidde-palletization throughput

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system attempts to automatically register unrecognized objects by performing additional scanning operations without requiring immediate manual intervention. The robotic arm can manipulate objects to present different surfaces to imaging devices, and the system processes these additional images to identify physical characteristics autonomously before pausing for manual registration only if automatic registration fails

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs more scanning and image processing operations than the minimum required, capturing multiple images from different angles and performing comprehensive analysis to ensure accurate identification. This excessive action reduces the need for manual intervention by providing redundant data for automatic registration algorithms

Inventive Principle:

Principle #16Partial or excessive action

3Measurement precision

If comprehensive scanning and registration procedures are implemented for all objects, then identification accuracy improves, but device complexity and processing time increase

Engineering Contradiction:

Improveobject recognition accuracyVSAvoidscanning and registration system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies different scanning and processing levels to different objects based on their recognition status. Recognized objects undergo minimal verification compared to unrecognized objects that require comprehensive scanning. The system focuses computational resources and imaging operations locally on objects that need them, rather than applying uniform complex procedures to all objects

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The de-palletization process is segmented into distinct phases: initial image capture, object recognition, physical characteristics identification, and manipulation. Each phase handles specific tasks independently, allowing the system to optimize processing for each segment and reduce overall complexity by breaking down the comprehensive scanning procedure into manageable, specialized sub-processes

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11501445B2Robotic system with automated package scan and registration mechanism and methods of operating the same

Publication Date: 2022.11.15 MUJIN INC

AI summary

A system and method for operating a robotic system to scan and register unrecognized objects is disclosed. The robotic system may use an image data representative of an unrecognized object located at a start location to implement operations for transferring the unrecognized object from the start location. While implementing the operations, the robotic system may obtain additional data, including scanning results of one or more portions of the unrecognized object not included in the image data. The robotic system may use the additional data to register the unrecognized object.