Mixed-Object Depalletizing with Constant Pallet Mode

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing depalletization systems struggle with unloading objects of different sizes and shapes from a pallet efficiently, leading to increased time and operational inefficiencies due to the reliance on time-consuming machine learning models for perception operations.

Innovation Solution

A depalletization system operates in a 'constant pallet mode' where, upon detecting similar dimensions of objects on the top layer, it maintains consistent settings and reduces the need for continuous image capture, using a perception subsystem to determine initial grasp poses and incrementally lifting objects based on stored data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If machine learning models are used for perception operations to handle mixed objects, then the system can identify and sort different sizes and shapes, but the cycle time increases significantly

Engineering Contradiction:
Improveability to handle mixed objectsVSAvoidcycle time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs perception operations only once at the beginning to capture an image of all objects on the pallet. This preliminary action provides sufficient information to plan the entire depalletization sequence without requiring continuous perception during execution, thereby reducing cycle time while maintaining adaptability to mixed objects

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the depalletization plan based on the initial perception data, modifying grasp poses and lift heights incrementally as objects are removed. This allows the system to handle mixed objects adaptively without repeating time-consuming perception operations

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If continuous image capture is performed to track object positions, then positioning accuracy is maintained, but the productivity decreases

Engineering Contradiction:
Improvepositioning accuracyVSAvoidproductivity
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The perception subsystem captures the complete arrangement of objects on the pallet in advance, storing this spatial information for reference during depalletization. This preliminary measurement provides sufficient positioning accuracy without requiring continuous image capture, thereby maintaining productivity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates a digital representation or model of the pallet layout from the initial image, using this copy to plan and execute depalletization operations. This eliminates the need for continuous physical measurement while maintaining positioning accuracy through computational geometry

Inventive Principle:
Principle #26Copying

3Manufacturing precision

If the system adapts to each unique object configuration, then handling precision is improved, but the operational complexity increases

Engineering Contradiction:
Improvehandling precisionVSAvoidoperational complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system implements dynamic replanning that incrementally adjusts the depalletization sequence based on initial perception data. Rather than requiring complex real-time adaptation for each object, the system dynamically modifies the plan as objects are removed, maintaining handling precision while reducing operational complexity

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses the initial perception image to automatically generate and adjust the depalletization plan without external intervention. The planning algorithm itself adapts to the specific object configuration, eliminating the need for complex external control mechanisms

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4173772B1Apparatuses and computer-implemented methods for depalletizing mixed objects
Publication Date: 2026.03.11 INTELLIGRATED HEADQUARTERS LLC
  • EP4173772B1 patent drawingFigure 1A
  • EP4173772B1 patent drawingFigure 1B
  • EP4173772B1 patent drawingFigure 2

AI summary

Methods, apparatuses, systems, computing devices, and/or the like are provided. An example method may include receiving, from a perception subsystem associated with an object depalletization system, first imaging data associated with a plurality of objects disposed on a pallet; calculate, based at least in part on the first imaging data, one or more comparative dimension measures associated with the plurality of objects; determine whether the one or more comparative dimension measures satisfy a comparative dimension threshold range; and in response to determining that the one or more comparative dimension measures satisfy the comparative dimension threshold range, cause an execution subsystem associated with the object depalletization system to operate in a constant pallet mode.