Automated Box Picking Optimization via Equivalence Classes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated systems for box picking and decanting face challenges such as positioning difficulties, inflexibility, and inefficiency due to reliance on classical image processing and machine learning methods, which require extensive training and are prone to overtraining or undertraining, making them costly and difficult to adapt to new items or configurations.
Innovation Solution
A method and system that utilize a multidimensional optimization approach to rank sequences of legal picks based on equivalence classes, employing a weighting vector and dot product calculations to optimize the number of picks required, allowing for efficient and flexible handling of box-like items without the need for extensive training or predetermined configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If classical image processing and machine learning methods are used for automated box picking, then the system can achieve automated operation, but the system requires extensive training and is prone to overtraining or undertraining, making it costly and difficult to adapt
Solution Approach 1:
The patent replaces classical image processing and machine learning systems with a multidimensional optimization approach. Instead of using neural networks and training data, the system uses mathematical optimization with equivalence classes and weighting vectors to determine pick sequences, eliminating the need for extensive training while maintaining automated operation.
Solution Approach 2:
The patent changes the fundamental parameters of the system by transitioning from data-driven machine learning parameters to optimization-based parameters. The system uses equivalence classes, weighting vectors, and dot product calculations to dynamically adapt to different box configurations without retraining, enabling both automation and adaptability.
2Extent of automation
If machine learning methods are used for box picking, then automated operation can be achieved, but the system is prone to overtraining or undertraining issues
Solution Approach 1:
The patent eliminates machine learning training processes entirely by substituting them with a deterministic optimization framework. The system uses equivalence classes and weighting vectors to reliably determine pick sequences without any training phase, ensuring consistent and reliable automated operation across different scenarios.
3Measurement precision
If extensive training is used to improve system performance, then automation accuracy can be improved, but the system becomes costly and difficult to adapt
Solution Approach 1:
The patent replaces complex machine learning training systems with a simpler optimization-based approach. By using equivalence classes and weighting vectors, the system achieves high picking accuracy through mathematical optimization rather than extensive training, reducing both time and computational complexity.
4Ease of manufacture
If the system uses predetermined configurations for box picking, then the picking process can be standardized, but the system becomes inflexible and difficult to adapt to new items
Solution Approach 1:
The patent makes the system dynamic by using optimization-based equivalence classes that can adapt to different box configurations in real-time. Instead of fixed predetermined configurations, the system dynamically determines optimal pick sequences based on current box arrangements, maintaining standardization through the optimization framework while achieving flexibility through adaptive calculations.
Data Source
AI summary
A method and system for quickly emptying a plurality of substantially identical target items from a transport structure with a picking tool for placement into a transport container are provided. The method includes the step of providing a plurality of equivalence classes which partition all possible sequences of legal picks. Each equivalence class containing a subset of items from a ranked configuration. The method also includes ranking the equivalence classes in order of expected time efficiency and selecting the best ranked equivalence class for picking.