Robotic Picking Training Using 3D Models and Center of Gravity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automated storage and retrieval systems face challenges in efficiently picking packages of varying size, weight, and orientation due to non-uniform weight distribution and random orientations of articles at the picking station.

Innovation Solution

A robotic package retrieval system that determines an optimal picking location by identifying the article, creating a 3D model, and determining the 3D center of gravity, which is then used to select the appropriate picking location, regardless of the article's weight distribution or orientation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a robotic manipulator picks articles with non-uniform weight distribution and random orientations, then the system can handle diverse packages, but the picking accuracy and efficiency deteriorate due to difficulty in determining optimal grasp points

Engineering Contradiction:
Improveability to handle diverse packagesVSAvoidpicking accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system performs preliminary actions by creating comprehensive 3D models and determining center of gravity locations for all articles before the actual picking operation. This advance preparation allows the robotic manipulator to quickly retrieve pre-calculated optimal grasp points from storage, eliminating the need for complex real-time calculations during picking operations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates digital 3D models that are copies of the physical articles, including their geometric features and weight distribution characteristics. These digital models allow the system to simulate and determine optimal grasp points without physically manipulating the actual articles, improving both accuracy and efficiency.

Inventive Principle:
Principle #26Copying

2Measurement precision

If the robotic system determines optimal picking locations through real-time 3D modeling and center of gravity calculation, then picking accuracy improves, but the processing time and system complexity increase

Engineering Contradiction:
Improvepicking accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs all complex 3D modeling and center of gravity calculations in advance, before the articles need to be picked. The optimal grasp points are determined and stored beforehand, allowing the robotic manipulator to simply retrieve this pre-calculated information during operation, dramatically reducing processing time while maintaining high accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system replaces real-time mechanical sensing and trial-and-error physical manipulation with pre-computed digital models and gravitational calculations. This substitution of computational methods for physical experimentation eliminates time-consuming iterative adjustments during the actual picking process.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Ease of operation

If the system creates detailed 3D models and determines center of gravity for each article, then the optimal picking location selection improves, but the computational resources and system complexity increase

Engineering Contradiction:
Improveoptimal picking location selectionVSAvoidsystem complexity
Core Design Contradiction:
Ease of operationVSDevice complexity

Solution Approach 1:

The system performs all complex computational tasks of 3D modeling and center of gravity determination in advance, storing the results for later retrieval. This shifts the computational burden from the operation phase to the preparation phase, making the actual picking operation simpler and more straightforward.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates digital 3D models as simplified representations of physical articles. These digital copies contain all necessary geometric and gravitational information, allowing the system to work with lightweight data structures rather than complex physical analysis during operation, reducing computational complexity.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3651944B1Robotic picking training technique
Publication Date: 2025.05.28 DEMATIC CORP
  • EP3651944B1 patent drawingFigure 1
  • EP3651944B1 patent drawingFigure 2
  • EP3651944B1 patent drawingFigure 3

AI summary

A method of training a robot to pick a plurality of articles includes determining an identity of an article and determining if the article has a uniform distribution of weight. If the article has a uniform distribution of weight, a three-dimensional (3D) model of the article is created and stored in a database record associated with the identity. If the article does not have a uniform distribution of weight, a 3D model is created, two-dimensional (2D) images of each side of the article are created, and a 3D center of gravity of the article is determined. The 3D model, the 2D images, and the 3D center of gravity of the article are stored in a database record associated with the identity of the article.