Robot Picking Simulation Using Realistic Bulk Pile Postures

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current robot simulation apparatuses fail to accurately reproduce the bulk pile state of workpieces during actual operations, often featuring unnatural postures and environments, leading to inaccurate picking motion simulations.

Innovation Solution

A robot simulation apparatus that sets workpiece models with specific posture conditions, generates bulk pile data excluding unnatural postures, and performs physical simulations to update data, ensuring simulations closely resemble actual operations, thereby improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If physical simulation of completely randomly falling workpiece models is performed, then the simulation covers all possible postures, but the bulk pile state does not approximate the actual operation environment

Engineering Contradiction:
Improveaccuracy of bulk pile state reproductionVSAvoidnaturalness of workpiece posture
Core Design Contradiction:
ReliabilityVSManufacturing precision

Solution Approach 1:

The patent changes the parameters of workpiece postures by setting minimum and maximum angle ranges that conform to actual operation environments. Instead of allowing completely random postures, the system constrains postures within realistic angular ranges, thereby improving the reliability of bulk pile state reproduction while maintaining naturalness.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies different quality constraints to different aspects of workpiece postures. By setting specific angle ranges for different orientations and positions, the system ensures that each local aspect of the posture conforms to actual operational realities, resulting in a more accurate overall bulk pile state.

Inventive Principle:
Principle #3Local quality

2Reliability

If posture restrictions are applied to workpiece models, then the bulk pile state approximates actual operation environment, but workpiece models in unnatural postures are excluded

Engineering Contradiction:
Improveaccuracy of bulk pile state reproductionVSAvoidcoverage of workpiece postures
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent defines specific angle range parameters (minimum and maximum angles) that represent actual operation environments. By changing the posture parameters to fall within these defined ranges, the system achieves accurate bulk pile state reproduction while maintaining adequate coverage of realistic workpiece postures.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent creates a virtual copy of the actual operation environment by reproducing the angular ranges and posture distributions observed in real operations. This copying approach ensures that the simulation accurately reflects actual conditions without requiring every possible unnatural posture to be included.

Inventive Principle:
Principle #26Copying

3Reliability

If multiple workpiece models with different postures are simulated, then the simulation comprehensively covers operation scenarios, but the simulation time and computational resources increase

Engineering Contradiction:
Improvecomprehensiveness of operation scenario coverageVSAvoidsimulation execution time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent changes the approach to posture simulation by using parameterized angle ranges instead of enumerating every possible posture. This allows the system to comprehensively cover operation scenarios within defined angular constraints while significantly reducing simulation time through efficient parameter sampling rather than exhaustive posture generation.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The solution enables more accurate simulation results by generating bulk pile data that reflects real-world operating environments, allowing for suitable setting adjustments and verifications without preparing actual workpieces, thus enhancing the reliability of picking motion simulations.

Implementation Method 1

an arm of a robot is moved to the grasping position, and an end effector such as a hand portion grasps the workpiece

Methodology Applied
Scientific EffectMechanical motion:

Implementation Method 2

a physical simulation of falling of a plurality of workpiece models in a virtual work space is performed in advance

Methodology Applied
Scientific EffectGravity: Gravitation

Data Source

PatentUS10913157B2Robot simulation apparatus and robot simulation method
Publication Date: 2021.02.09 KEYENCE CORP
  • US10913157B2 patent drawing
  • US10913157B2 patent drawing
  • US10913157B2 patent drawing

AI summary

The robot simulation apparatus includes: a workpiece model setting unit 11 that sets a workpiece model; a posture condition setting unit 16 that sets a condition related to a posture taken by the workpiece model, as a posture condition, when the workpiece model set by the workpiece model setting unit 11 is disposed in a virtual work space as a virtually formed work space; a bulk pile data generating unit 20 that generates bulk pile data of the plurality of workpiece models piled up in the virtual work space, in accordance with the posture condition set by the posture condition setting unit 16; and a picking motion simulating unit 30 that executes a picking motion simulation for verifying the picking motion of the workpiece models in the virtual work space, with respect to the bulk pile data generated by the bulk pile data generating unit 20.