3D Vision Segmentation for Faster Robotic Workpiece Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robotic vision systems process large image data sets, often requiring more memory and processing power, which can slow down robotic operations and increase costs, necessitating a more efficient approach to handle 3D image data for robotic applications.

Innovation Solution

A robotic system equipped with a 3D vision sensor and a control system that filters 3D image data by segmenting the image into a region of interest and storing only that data, allowing the robot to perform work based on the filtered image, thereby reducing the data set size and improving processing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the vision system processes the complete 3D image data set, then comprehensive scene information is obtained, but memory and processing requirements increase and operational speed decreases

Engineering Contradiction:
Improvescene informationVSAvoidoperational speed
Core Design Contradiction:
Loss of informationVSProductivity

Solution Approach 1:

The patent segments the 3D image data into multiple point clouds based on depth ranges. Each point cloud contains only the points within a specific depth range, allowing the system to process smaller subsets of data independently. This segmentation enables the robot to focus computational resources on relevant regions while maintaining comprehensive scene understanding across multiple segments.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the necessary portion of the 3D image data corresponding to the robot's workspace and relevant objects. By identifying and extracting point clouds within the robot's reachable depth range, the system eliminates unnecessary data processing while preserving all information needed for task execution.

Inventive Principle:
Principle #2Taking out (Extraction)

2Power

If a larger memory and faster processor are used to handle complete image data, then processing capability is improved, but system cost increases

Engineering Contradiction:
Improveprocessing capabilityVSAvoidsystem cost
Core Design Contradiction:
PowerVSDevice complexity

Solution Approach 1:

The patent applies partial action by processing only the necessary portion of the 3D image data - specifically, point clouds within the robot's workspace depth range. Rather than processing the complete scene data, the system performs sufficient processing on relevant subsets, reducing memory and computational requirements while maintaining adequate processing capability for robotic tasks.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If the complete 3D image data is processed, then accurate workpiece location is achieved, but data processing time increases

Engineering Contradiction:
Improveworkpiece location accuracyVSAvoiddata processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the 3D image into multiple depth-based point clouds and processes only those segments relevant to the robot's workspace. This segmentation allows the system to maintain accurate workpiece location by focusing computational effort on relevant depth ranges, significantly reducing processing time compared to analyzing the complete image data set.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary filtering of the 3D image data to identify and extract only the point clouds within the robot's reachable depth range before detailed analysis. This preliminary action reduces the data volume requiring intensive processing, thereby decreasing overall processing time while preserving location accuracy for relevant objects.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3250346B13D segmentation for robotic applications
Publication Date: 2021.08.18 ABB (SCHWEIZ) AG
  • EP3250346B1 patent drawingFigure 1
  • EP3250346B1 patent drawingFigure 2
  • EP3250346B1 patent drawingFigure 3A~3B

AI summary

A robotic system includes a robot having an associated workspace; a vision sensor constructed to obtain a 3D image of a robot scene including a workpiece located in the workspace; and a control system communicatively coupled to the vision sensor and to the robot. The control system is configured to execute program instructions to filter the image by segmenting the image into a first image portion containing substantially only a region of interest within the robot scene, and a second image portion containing the balance of the robot scene outside the region of interest; and by storing image data associated with the first image portion. The control system is operative to control movement of the robot to perform work, on the workpiece based on the image data associated with the first image portion.