Virtual Robot Simulation Feedback for Real-World Motion Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing simulation systems for industrial robots face challenges in accurately replicating the operations of actual robots due to differences in characteristics such as inertia, rigidity, and viscosity, leading to discrepancies between simulated and actual robot performances.

Innovation Solution

A simulation system that includes processing circuitry and storage, capable of outputting image data of a virtual robot model performing target operations, with a simulator functional part that adjusts and reflects changes in operations to match the actual robot's characteristics, allowing for improved alignment between simulated and actual robot behaviors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If a virtual robot model is used for simulation, then control program development becomes possible before actual robot deployment, but operational differences arise due to characteristic discrepancies between virtual and actual robots

Engineering Contradiction:
Improvecontrol program development timeVSAvoidoperation accuracy
Core Design Contradiction:
Loss of timeVSReliability

Solution Approach 1:

The patent performs preliminary actions by acquiring actual robot operation data before simulation, using it to correct the virtual robot model's characteristics. This preliminary correction ensures that the virtual model more accurately reflects the actual robot's behavior, reducing operational differences while enabling early control program development.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by comparing operation data from the actual robot with operation data from the virtual robot model. The differences detected through this feedback mechanism are used to correct the virtual model's characteristics, creating a continuous improvement loop that reduces operational discrepancies.

Inventive Principle:
Principle #23Feedback

2Measurement precision

If the virtual robot model characteristics are corrected to match actual robot characteristics, then operational accuracy improves, but the complexity of the simulation system increases

Engineering Contradiction:
Improveoperation similarityVSAvoidsimulation system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent uses copying by acquiring operation data from the actual robot and applying it to correct the virtual robot model. This copying approach allows the virtual model to replicate actual robot characteristics without requiring complex physical modifications or additional hardware, maintaining system simplicity while improving accuracy.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent applies parameter changes by modifying the virtual robot model's characteristics based on actual robot operation data. This involves adjusting parameters such as inertia, rigidity, and viscosity to match actual measurements, thereby improving operational accuracy through systematic parameter optimization rather than structural complexity.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20240037294A1Simulation apparatus and simulation system
Publication Date: 2024.02.01 KAWASAKI JUKOGYO KK
  • US20240037294A1 patent drawing
  • US20240037294A1 patent drawing
  • US20240037294A1 patent drawing

AI summary

A simulator includes circuitry and a storage. The storage stores target operation data indicative of a series of target operations, and data relevant to a virtual robot model. The processing circuitry performs: causing the robot model to operate according to the target operation data; outputting image data of the operating robot model to a display; accepting an input of a change in a first operation that is operation of the robot model according to the target operation data and is displayed on the display; associating information indicative of a state of the robot model in a second operation, to which the change is reflected, with information on the first operation included in the target operation data; and storing the associated information in the storage. The circuitry outputs image data indicative of the second operation to the display, when causing the robot model to perform the first operation.