Product Handling Control Using Geometric-Only Training Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automation systems for handling products require a high individual modification and programming effort to create training data sets, especially when considering the variety of products and their starting positions, which increases technical effort and processing time.
Innovation Solution
The method involves generating training data sets exclusively from geometric data contained in product image data using a computer program, allowing for the creation of a statistical model that controls the handling device without considering other physical parameters, thus simplifying the application of artificial intelligence for handling tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If training data sets are created using simulation with computer for handling tasks, then the automation system can handle various products and positions, but the individual modification and programming effort is very high
Solution Approach 1:
The patent uses image data from cameras to create simplified geometric representations (2D/3D models) of products instead of creating detailed physical simulations. These geometric copies capture essential shape and position information needed for handling while eliminating the need for complex physical property simulations, thereby reducing programming effort while maintaining adaptability to various products
Solution Approach 2:
The patent extracts only the necessary geometric information from product images, separating it from other physical parameters like weight, material properties, and dynamics. This extraction focuses training data creation on shape and position data only, which are sufficient for gripping and positioning tasks, thus reducing the complexity and programming effort required while preserving versatility
2Reliability
If comprehensive simulation is used to account for product weight and movement dynamics, then gripping tasks are fulfilled securely, but large data amounts have to be processed increasing technical effort and time
Solution Approach 1:
The patent extracts and uses only geometric data from product images, deliberately excluding weight and dynamic properties from the training data. This extraction approach reduces the volume of data to be processed while maintaining sufficient reliability for gripping and positioning tasks, as these geometric features are the critical factors for successful handling operations
Solution Approach 2:
The patent applies partial simulation by focusing only on the essential geometric aspects needed for handling tasks rather than performing complete physical simulations. This partial approach processes less data faster while still achieving reliable gripping and positioning by concentrating on the most critical parameters (shape and position) rather than all physical properties
Data Source
AI summary
The invention relates to a method for handling products (17) using an automation system and to an automation system (10), the products being captured by means of an imaging sensor (18) of a control device (12) of the automation system and being handled by means of a handling mechanism (13) of a handling device (11) of the automation system, the control device processing sensor image data from the imaging sensor and controlling the handling device as specified by training data sets contained in a data memory (21) of the control device, the training data sets comprising training image data and/or geometric data and control instructions associated therewith, the training data sets being generated, as a statistical model, exclusively from geometric data contained in the training image data of products, by means of a computer using a computer program product executed thereon, the training data sets being transmitted to the control device.