Robot Work Program Transfer With ML-Based Update Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing industrial robot systems face complexity in updating control programs as the types and amount of data increase, leading to potential improper program imports and decreased efficiency when work diversification occurs.

Innovation Solution

A program transfer system that includes a communication unit, determination unit, program import unit, collection unit, and program update unit, which communicates with robots, checks and updates work programs using machine learning based on collected work data to ensure accurate and efficient program imports and updates.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the work for the robot is diversified, then the adaptability of the robot system is improved, but the control program complexity increases

Engineering Contradiction:

Improvework diversificationVSAvoidcontrol program complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the control program into multiple versions corresponding to different work types. Each work type has its own dedicated control program version, allowing the robot to switch between specialized programs rather than managing one complex universal program. This segmentation reduces the complexity within each program while maintaining overall system adaptability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by automatically determining and importing the necessary control program version before the robot begins a new work type. The system proactively selects and loads the appropriate program version based on the intended work, preventing the accumulation of complexity and ensuring the robot is ready with the correct specialized program before execution.

Inventive Principle:

Principle #10Preliminary action

2Adaptability or versatility

If the amount of control program data increases, then the functionality of the robot is improved, but the update work becomes more complicated

Engineering Contradiction:

Improvecontrol program functionalityVSAvoidupdate work complexity

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The control program is segmented into multiple versions stored in the storage unit, with each version corresponding to specific work types or purposes. This segmentation allows updates to be applied to individual program versions independently, rather than updating one large monolithic program, thereby simplifying the update process while maintaining comprehensive functionality.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The determination unit uses feedback mechanisms to automatically identify which control program version should be imported based on the robot's current work context. This automated determination reduces manual intervention in the update process, making it easier to manage multiple program versions and reducing update work complexity.

Inventive Principle:

Principle #23Feedback

3Productivity

If automatic program import is implemented, then the productivity of program updates is improved, but the risk of improper program import increases

Engineering Contradiction:

Improveprogram update efficiencyVSAvoidprogram import accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The determination unit implements feedback by continuously monitoring the robot's work context and automatically determining the appropriate control program version to import. This closed-loop approach ensures that the correct program is selected based on actual work requirements, maintaining high productivity while minimizing improper imports through automated decision-making based on real-time feedback.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The determination unit acts as an intermediary between the robot's work requirements and the control program selection process. It mediates the automatic import process by intelligently determining which program version should be imported, thereby maintaining productivity while ensuring accuracy through this intermediate decision-making layer that prevents improper program imports.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11237819B2Program transfer system performing machine learning for preceding robotic work data and robot system having the program transfer system

Publication Date: 2022.02.01 SEIKO EPSON CORP

AI summary

A program transfer system includes: a communication unit communicating with a robot and outputting a result of the communication; a determination unit checking a work program provided in the robot, based on the result of the communication, and determining whether to import a different target work program from the work program or not; a program import unit importing the target work program to the robot when the determination unit determines the importing; a collection unit collecting work data of complete work from the robot; and a program update unit performing machine learning using the work data collected by the collection unit, and updating the target work program, based on a result of the machine learning.