Package Registration Using Minimum Viable Regions for De-Palletizing

Solution Overview

Problem

Current robotic systems face challenges in accurately registering and handling packages on pallets, particularly when physical characteristics such as dimensions and weight are unknown, leading to potential mishandling and increased risk of injury to human workers during de-palletization.

Innovation Solution

A robotic system that uses 2D and 3D imaging data to identify exposed edges and corners of packages, generates Minimum Viable Regions (MVRs) to estimate the surface area required for lifting, and verifies these regions through expansion and reduction processes to ensure accurate registration and handling of unrecognized packages.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a robotic system uses image matching with pre-registered packages, then the system can operate with known physical characteristics, but the system fails when encountering unrecognized packages leading to handling errors

Engineering Contradiction:

Improvepackage handling reliabilityVSAvoidability to handle unrecognized packages

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The robotic system performs self-registration by autonomously capturing images of unrecognized packages, extracting physical characteristics through image processing, and storing this data in a registration data source. This self-service capability allows the system to adapt to new package types without human intervention, resolving the contradiction between reliability for known packages and adaptability for unrecognized packages.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary registration actions by capturing images and extracting physical characteristics of packages before they are handled. This preliminary action creates a database of registered packages, enabling the robotic arm to reliably handle future packages by comparing them against this pre-established registration data.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If manual registration of packages is performed, then accurate physical characteristics can be obtained, but human intervention increases time consumption and reduces productivity

Engineering Contradiction:

Improvephysical characteristics accuracyVSAvoidde-palletization throughput

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system replaces manual human registration with an automated imaging and image processing system. The imaging device captures package images, and software algorithms automatically extract physical characteristics such as dimensions, weight, and center of mass. This substitution eliminates human intervention while maintaining measurement precision, thereby increasing productivity during de-palletization operations.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The robotic system performs self-registration by autonomously capturing images of unrecognized packages, extracting physical characteristics through image processing, and storing this data in a registration data source. This self-service capability allows the system to adapt to new package types without human intervention, resolving the contradiction between reliability for known packages and adaptability for unrecognized packages.

Inventive Principle:

Principle #25Self-service

3Adaptability or versatility

If the robotic arm handles packages without knowing their physical characteristics, then the system can attempt to process any package, but mishandling occurs especially with heavy or lop-sided packages

Engineering Contradiction:

Improveability to process any packageVSAvoidsafe package handling

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system performs preliminary registration actions by capturing images and extracting physical characteristics of packages before they are handled. This preliminary action creates a database of registered packages, enabling the robotic arm to reliably handle future packages by comparing them against this pre-established registration data.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from image matching to determine whether a package is recognized or unrecognized. For unrecognized packages, the system performs registration and updates the database. This feedback loop ensures that the robotic arm always has access to accurate physical characteristics for safe and reliable package handling.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS11189033B2Robotic system with automated package registration mechanism and auto-detection pipeline

Publication Date: 2021.11.30 MUJIN INC

AI summary

The present disclosure relates to detecting and registering unrecognized or unregistered objects. A minimum viable range (MVR) may be derived based on inspecting image data that represents objects at a start location. The MVR may be determined to be a certain MVR or an uncertain MVR according to one or more features represented in the image data. The MVR may be used to register corresponding objects according to the certain or uncertain determination.