Robotic Singulation Weight Detection for Double-Pick Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated robotics face challenges in singulating individual objects from a collection, especially when dealing with diverse object types, as existing AI solutions require extensive training and are prone to errors, making it difficult to accurately determine if objects have been successfully separated.
Innovation Solution
A singulation system comprising a container, a robotic manipulator, and a sensor to measure the weight of objects, with a computer system comparing the measured weight to an acceptable range to confirm successful singulation, using RFID tags and RF signals for object identification and a tilt tray or conveyor belt for object handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If AI technologies are used to optically recognize objects, then object recognition capability is improved, but training time and computational resources increase significantly
Solution Approach 1:
The patent extracts the essential identifying feature (weight) from the complex object recognition problem. Instead of using AI to analyze multiple visual features requiring extensive training, the system simply measures weight, which directly identifies object type without needing to learn from training data.
Solution Approach 2:
The patent replaces the optical/AI recognition system with a mechanical weighing system. This substitution eliminates the need for complex AI training while providing reliable object identification through direct physical measurement.
2Measurement precision
If AI solutions are used for object recognition, then object identification accuracy is improved, but system complexity increases
Solution Approach 1:
The patent extracts the critical identifying characteristic (weight) from the complex set of visual features that AI would need to process. This reduction to a single measurable parameter dramatically simplifies the system while maintaining identification accuracy.
Solution Approach 2:
The patent uses simple, inexpensive weight sensors instead of complex AI systems. These basic sensing elements provide reliable identification without the computational complexity and resource requirements of AI models.
3Extent of automation
If robotic manipulator picks objects from collection, then automation is improved, but ability to detect multiple objects (double pick) worsens
Solution Approach 1:
The patent implements feedback by measuring the weight of each picked object and comparing it against expected weight ranges. This feedback mechanism reliably detects when multiple objects are picked together (double pick) or when picking fails, enabling real-time correction.
Solution Approach 2:
The patent replaces complex vision-based detection systems with simple weight measurement for detecting multiple objects. This mechanical approach provides more reliable detection of singulation failures than optical methods.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively singulates objects by accurately determining if they have been separated, reducing errors and improving the efficiency of processing diverse object types, enhancing the automation of industrial operations.
Implementation Method 1
a sensor configured to measure a weight of the individual object
Implementation Method 2
using RFID tags and RF signals for object identification
Data Source
AI summary
A system for singulating objects includes a bin for receiving a collection of objects, a robotic manipulator for grasping objects from the bin, a scale for measuring a weight of the grasped objects, and a computer system for comparing measured weights to acceptable weight ranges to detect double picks. Methods include determining acceptable weight ranges by weighing a plurality of objects.