Robot Learning From Worker Force and Position for Faster Setup
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automated manufacturing systems requiring robot manipulators often need specialized workers for setup, leading to inefficiencies due to awkward operations and high workload, resulting in low robot work efficiency.
Innovation Solution
A robot system that includes a learning apparatus using machine learning to obtain a learned model based on force, position, and workpiece information, allowing the robot to mimic worker operations and improve efficiency without requiring specialized knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a worker with specialized knowledge operates the robot manipulator to set up the automated manufacturing system, then the robot can be configured to perform work, but the setup process becomes complex and time-consuming
Solution Approach 1:
The system records the worker's operation data (position, force, acceleration) during manual handling of workpieces and creates a learned model that copies this behavior. The robot then reproduces the worker's operations automatically based on this learned model, eliminating the need for complex specialized setup procedures while maintaining the effectiveness of expert operations
Solution Approach 2:
The system replaces the mechanical knowledge-based setup process with a data-driven machine learning approach. Instead of requiring workers to manually program and configure the robot through complex interfaces, the system captures operational data and uses machine learning algorithms to automatically generate control parameters, substituting human expertise with an automated learning system
2Productivity
If a worker manually operates the robot manipulator, then the robot can be taught to perform tasks, but the operation becomes awkward and inefficient
Solution Approach 1:
The system incorporates acceleration information and force data as feedback parameters to continuously monitor and adjust the robot's operations. By comparing the robot's actual performance with the learned model derived from worker operations, the system optimizes control parameters to eliminate awkward movements and improve operational smoothness while maintaining high productivity
Solution Approach 2:
The system dynamically adjusts robot operations by incorporating acceleration information and force data into the control framework. The learned model captures the dynamic characteristics of manual operations and enables the robot to adapt its movements in real-time, transforming rigid automated operations into smooth, natural-looking motions that maintain high efficiency
3Reliability
If specialized workers are required to set up and operate the automated manufacturing system, then the system can function properly, but the workload and time requirements increase
Solution Approach 1:
The system performs self-learning by automatically capturing operation data from workers and generating its own control parameters through machine learning. The robot autonomously develops its own learned model without requiring specialized workers to manually program or configure it, eliminating time-consuming setup procedures while ensuring reliable system functionality through data-driven optimization
Data Source
AI summary
A robot system includes a robot, and an information processing portion. The information processing portion is configured to obtain a learned model by learning first force information about a force applied by a worker to a workpiece, first position information about a position of a first portion of the worker, and first workpiece information about a state of the workpiece, and control the robot on a basis of output data of the learned model.