AI Bin-Picking Vision for Reliable Object Pose Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic devices for bin-picking face challenges in identifying objects due to variability in feature extraction methods and sensitivity to external environment changes, which affects their accuracy and reliability.
Innovation Solution
An electronic device equipped with a memory containing a training database based on an AI algorithm, sensors, and a processor that acquires data on objects using sensors, identifies location and positioning information, and transmits control signals to a picking tool for object retrieval, utilizing AI algorithms for improved object identification and handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional feature extraction methods are used to identify objects, then the device can perform bin-picking operations, but the identification accuracy deteriorates due to sensitivity to external environment changes and variability in feature extraction schemes
Solution Approach 1:
The patent replaces traditional mechanical feature extraction methods with an artificial intelligence-based vision system. The AI algorithm processes images captured by sensors to identify objects, their positions, and orientations, substituting the conventional mechanical approach with intelligent data processing that is insensitive to environmental variations.
Solution Approach 2:
The patent changes the parameters used for object identification from fixed geometric features to AI-learned features that adapt to different conditions. The system uses trained neural networks to extract meaningful parameters from images, allowing accurate identification despite changes in lighting, object arrangement, or environmental conditions.
2Adaptability or versatility
If multiple feature extraction schemes are used to handle different objects, then the device can identify various objects, but the device complexity increases due to the need to vary extraction schemes according to object types
Solution Approach 1:
The patent implements a universal AI-based feature extraction system that can handle multiple object types with a single unified approach. The neural network is trained on diverse data and can automatically adapt to different objects, eliminating the need for multiple specialized extraction schemes and simplifying the overall system architecture.
Solution Approach 2:
The AI system performs self-learning and automatic adaptation to different objects without requiring manual configuration or switching of extraction schemes. The neural network automatically adjusts its parameters and features based on the input data, enabling the system to handle various objects autonomously with a single unified process.
Data Source
AI summary
According to various embodiments of the present invention, an electronic device comprises: a memory including instructions and a training database, which includes data, on at least one object, acquired on the basis of an artificial intelligence algorithm; at least one sensor; and a processor connected to the at least one sensor and the memory, wherein the processor can be configured to execute the instructions in order to acquire data on a designated area including the at least one object by using the at least one sensor, identify location information and positioning information on the at least one object on the basis of the training database, and transmit a control signal for picking the at least one object to a picking tool related to the electronic device on the basis of the identified location information and positioning information.