Automated Package Registration for Robotic Gripping of Unrecognized Loads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robotic systems face challenges in autonomously registering and handling unrecognized objects, leading to potential mishandling and increased manual intervention, especially with heavy or unevenly packed packages, which can result in safety risks and operational inefficiencies.
Innovation Solution
A robotic system that uses imaging sensors and depth maps to identify unrecognized objects by comparing sensor data to master registration data, derives minimum viable regions for gripping, and performs initial lifts to separate objects, allowing for autonomous registration and handling without human operator input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If human workers perform de-palletizing operations, then flexibility in handling various packages is improved, but labor costs increase and safety risks arise from bodily injuries
Solution Approach 1:
The robotic system autonomously performs de-palletizing operations without human intervention. The system self-registers unrecognized packages by capturing images, comparing them against a library, and automatically adjusting handling parameters, thereby eliminating the need for human workers while maintaining operational flexibility and safety
Solution Approach 2:
The patent replaces human mechanical operations with an automated robotic system equipped with imaging devices and control algorithms. The robotic arm with gripper substitutes human hands, while the image recognition system substitutes human visual identification, eliminating safety risks associated with manual handling
2Productivity
If industrial robots perform de-palletizing operations without automated registration, then operational speed is improved, but mishandling of heavy or lop-sided packages occurs
Solution Approach 1:
The system performs preliminary registration of packages by capturing images and determining physical characteristics (weight, dimensions, center of mass) before the robotic arm attempts to handle them. This preliminary identification ensures that the robot is fully informed about package properties prior to gripping and lifting, preventing mishandling while maintaining operational speed
Solution Approach 2:
The system uses image recognition feedback to continuously identify and register package characteristics. The control system compares captured images against the package library, receives feedback on recognition accuracy, and automatically adjusts handling parameters to ensure reliable identification of physical characteristics for safe manipulation
3Measurement precision
If manual registration of packages is performed, then accuracy in identifying physical characteristics is improved, but operational interruptions increase due to stoppages
Solution Approach 1:
The patent replaces manual registration operations with an automated image recognition system. The robotic system captures images of packages, processes them through computer vision algorithms, and automatically determines physical characteristics without human intervention, thereby maintaining measurement precision while eliminating operational stoppages
Solution Approach 2:
The system performs package registration and identification continuously as packages move through the de-palletizing process. The image capture and analysis occur in real-time without interrupting the robotic arm's operations, ensuring both accurate identification and uninterrupted workflow
4Productivity
If the robotic system attempts to handle unrecognized packages without initial registration, then operational continuity is improved, but mishandling and safety risks increase
Solution Approach 1:
The system performs preliminary image capture and registration for unrecognized packages before the robotic arm attempts to handle them. This preliminary identification step occurs quickly and automatically, ensuring that physical characteristics are known prior to manipulation, thereby preventing mishandling while maintaining operational continuity
Solution Approach 2:
The robotic system autonomously handles unrecognized packages by self-registering them through image capture and automatic comparison against the package library. The system determines physical characteristics without human assistance and adjusts its handling approach accordingly, ensuring safe operation while maintaining continuous productivity
Data Source
AI summary
A method for operating a robotic system includes obtaining and processing first data representative of an object at a start location. An event may be detected while implementing an operation based on the image data. A gripper height that corresponds to the event may be determined. Accordingly, the method may include calculating an object height that represents a height estimate of the object.


