Robotic Palletizing of Mixed Items Using Vision and Sensor Snugging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for palletizing and depalletizing heterogeneous items are inefficient and prone to instability due to the variety of item sizes, weights, and types, often requiring manual intervention and relying on human judgment, which can lead to unstable stacks and item damage.
Innovation Solution
A robotic system equipped with 3D cameras, force sensors, and a library of item types and grasp strategies uses programmable algorithms to identify and stack items based on their attributes, adjusting plans dynamically to ensure stability and efficiency, with human intervention available if needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual stacking by human workers is used, then item selection and placement can be done with human judgment, but the process is slow and inconsistent in ensuring stability
Solution Approach 1:
The patent replaces manual human stacking with an automated robotic system that uses computer vision (3D cameras) and algorithms to identify items and determine optimal placement. The robotic arm with gripper mechanically picks and places items based on computed stability criteria, eliminating human judgment while maintaining or improving stack stability through systematic evaluation of item attributes and stacking strategies.
Solution Approach 2:
The system enables self-service by allowing the robotic system to autonomously select items from the conveyor, determine their attributes through 3D imaging, compute optimal placement locations using stability algorithms, and execute the stacking without human intervention. The system serves itself by integrating perception, decision-making, and execution in one automated loop.
2Productivity
If items are stacked quickly without careful selection, then productivity increases, but the palletized stack becomes unstable and items may be damaged
Solution Approach 1:
The system performs preliminary actions by using 3D cameras to scan and identify item attributes (size, shape, weight estimation) before the actual stacking decision is made. The algorithm pre-computes the optimal placement location based on these attributes and stability criteria, so that when the robotic arm picks the item, the placement decision is already optimized for stability, enabling fast execution without compromising stack integrity.
Solution Approach 2:
The system incorporates feedback by using 3D vision to continuously monitor item attributes and stack configuration, then feeding this information back to the stacking algorithm which adjusts placement decisions in real-time. The system adapts to the actual items present on the conveyor and modifies stacking strategies based on observed item variations, ensuring stability even with heterogeneous items.
3Productivity
If a robotic system is used to automate palletizing, then productivity increases, but the system complexity increases due to variety of items and feed mechanisms
Solution Approach 1:
The patent implements universality by designing a robotic system with a generic 3D vision system and adaptive algorithm that can handle multiple item types and feed mechanisms through a unified approach. The stacking algorithm evaluates item attributes and applies general stability criteria that work across different item geometries and materials, rather than requiring item-specific programming. The robotic arm with adjustable gripper can accommodate various item sizes and shapes, reducing overall system complexity despite handling diverse items.
Solution Approach 2:
The system manages complexity by changing parameters dynamically - using 3D vision to measure item dimensions, orientation, and position, then adjusting stacking parameters (placement location, orientation, layer configuration) based on these measurements. The algorithm adapts stacking strategies by modifying parameters such as item orientation angles, placement coordinates, and gripper force based on real-time item characteristics, enabling flexible handling without hardcoding for each item type.
4Manufacturing precision
If 3D cameras and sensors are used to identify item attributes, then stacking accuracy improves, but the measurement and detection difficulty increases
Solution Approach 1:
The patent replaces complex manual measurement and inspection with automated 3D vision systems that use cameras and sensors to capture item geometry, position, and orientation. Computer vision algorithms automatically process the visual data to extract item attributes, eliminating the need for manual measurement tools or complex mechanical probing systems. The system achieves high measurement precision through optical methods rather than mechanical contact, simplifying the detection process despite handling diverse item types.
Data Source
AI summary
Techniques are disclosed to use a robotic arm to palletize or depalletize diverse items. In various embodiments, data associated with a plurality of items to be stacked on or in a destination location is received. A plan to stack the items on or in the destination location is generated based at least in part on the received data. The plan is implemented at least in part by controlling a robotic arm of the robot to pick up the items and stack them on or in the receptacle according to the plan, including by for each item: using one or more first order sensors to move the item to a first approximation of a destination position for that item at the destination location; and using one or more second order sensors to snug the item into a final position.