Robot Picking Control With 3D Learning for Bulk-Loaded Workpieces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional robot systems require human intervention and trial-and-error to optimize the picking up of workpieces in bulk-loaded states, leading to a low success rate due to the need for presetting workpiece extraction settings and operation programming.
Innovation Solution
A machine learning device that includes a state variable observation unit, an operation result obtaining unit, and a learning unit to learn manipulated variables for commanding a robot to perform picking operations, allowing for autonomous optimization of picking operations without human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If presetting workpiece extraction settings and programming robot operations manually, then the robot system can perform picking operations, but the success rate is low and requires continuous human intervention and trial-and-error optimization
Solution Approach 1:
The robot system performs self-learning by automatically observing picking operation results and refining its own control parameters without human intervention. The learning unit enables the system to improve its picking success rate through autonomous trial-and-error learning, where the robot adjusts its extraction settings and operation parameters based on observed outcomes, ultimately achieving high-success-rate operations independently.
Solution Approach 2:
The system implements a feedback mechanism where the learning unit receives picking operation results and uses this information to automatically adjust and optimize control parameters. The observed outcomes feed back into the learning process, enabling continuous improvement of the robot's picking performance without requiring manual reprogramming or intervention.
2Reliability
If manual trial-and-error optimization is performed to improve picking success rate, then operation parameters can be refined, but significant human time and effort are consumed
Solution Approach 1:
The robot system autonomously performs the optimization process that would otherwise require human operators. The learning unit automatically analyzes picking results and adjusts parameters, transferring the time-consuming optimization task from human operators to the automated system, thereby eliminating human time loss while maintaining continuous improvement of success rate.
Solution Approach 2:
The system performs preliminary learning and parameter optimization automatically during operation setup, rather than requiring manual trial-and-error later. By pre-learning optimal parameters through autonomous observation and adjustment, the system prepares itself in advance for high-success-rate operations, saving significant human time that would otherwise be spent on iterative optimization.
3Adaptability or versatility
If conventional programming methods are used to control robot picking operations, then basic functionality is achieved, but the system cannot adapt to random and bulk-loaded workpiece arrangements
Solution Approach 1:
The system transitions from static pre-programmed operations to dynamic adaptive learning. The learning unit enables the robot to continuously adjust its control parameters based on observed picking results, allowing it to adapt to various workpiece arrangements (random, bulk-loaded, etc.) without requiring separate programming for each scenario. This dynamic adaptation capability handles diverse arrangements uniformly.
Solution Approach 2:
The learning unit provides a universal solution that handles multiple workpiece arrangement types (random, bulk-loaded, organized) through a single adaptive system. Rather than requiring separate programming for each arrangement type, the system learns and adapts to handle all variations uniformly, simplifying the overall programming complexity while expanding adaptability across different scenarios.
Data Source
AI summary
A machine learning device that learns an operation of a robot for picking up, by a hand unit, any of a plurality of workpieces placed in a random fashion, including a bulk-loaded state, includes a state variable observation unit that observes a state variable representing a state of the robot, including data output from a three-dimensional measuring device that obtains a three-dimensional map for each workpiece, an operation result obtaining unit that obtains a result of a picking operation of the robot for picking up the workpiece by the hand unit, and a learning unit that learns a manipulated variable including command data for commanding the robot to perform the picking operation of the workpiece, in association with the state variable of the robot and the result of the picking operation, upon receiving output from the state variable observation unit and output from the operation result obtaining unit.