Method for gripping textile objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated laundry systems face challenges in reliably gripping textile objects, particularly flat linen, due to unpredictable orientations and sizes, which can lead to inaccurate identification of gripping points by imaging devices.
Innovation Solution
The method employs multiple imaging devices positioned around the textile object to capture images from various angles, ensuring that the gripping point can be determined accurately and reliably, regardless of the object's orientation, by comparing and evaluating the image data to guide the handling device's gripping mechanism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single imaging device is used to capture textile objects, then the device complexity is low, but the measurement precision of gripping point location deteriorates due to unreliable detection of arbitrarily oriented textile objects
Solution Approach 1:
The imaging task is segmented across multiple imaging devices, each capturing a specific viewing angle or direction of the textile object. This segmentation allows comprehensive coverage of all possible orientations without requiring a single complex imaging system, thereby improving measurement precision while managing device complexity through modular deployment
Solution Approach 2:
The system transitions from single-view (2D) imaging to multi-view (3D spatial) imaging by positioning imaging devices at different locations and angles. This dimensional expansion enables accurate detection of gripping points regardless of the textile object's orientation in space, resolving the contradiction between measurement precision and device complexity
2Reliability
If multiple imaging devices are deployed to capture textile objects from different angles, then the reliability of gripping point detection improves, but the device complexity increases
Solution Approach 1:
Each imaging device is designed with universal functionality to capture textile objects from its specific viewing angle. The system uses multiple identical or similar imaging devices rather than complex specialized systems, improving reliability through redundancy and comprehensive coverage while keeping individual device complexity low and manageable
Solution Approach 2:
The system processes image data from multiple imaging devices to identify the optimal viewing angle and select the most suitable image for gripping point detection. This feedback mechanism ensures reliable automatic gripping by dynamically choosing the best data source, thereby improving reliability without requiring all imaging devices to operate at maximum complexity simultaneously
3Adaptability or versatility
If imaging devices capture only specific portions of textile objects, then the device complexity remains low, but the adaptability to handle textile objects of different sizes deteriorates
Solution Approach 1:
Each imaging device is configured to capture a specific local portion or viewing angle of the textile object, optimized for detecting gripping points in that region. This local quality approach allows the system to handle various sizes by selecting appropriate local views, maintaining adaptability while keeping individual device configurations simple and manageable
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In laundries, items of laundry (14) need to be picked up at various points. This is labor-intensive. Therefore, efforts are being made to automate the picking of laundry items (14). Currently, a single camera (43) is used for this purpose. The invention provides for the automatic picking of laundry items (14) using multiple cameras (43) by having several cameras (43) simultaneously capture image data of the respective laundry item (14) to be picked up. This ensures that the cameras (43) always provide usable image data of the laundry item (14), even if, for example, a camera (43) is temporarily obscured, defective, or if the viewing direction (44) of a camera is not directed at the surface of the laundry item (14). Due to the multiple cameras (43), image data of the laundry item (14) can always be generated, enabling a handling device (11) to reliably pick up the laundry item (14) automatically.