Workpiece Recognition Switching Between 2D Edge and 3D Matching

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing technologies face difficulties in detecting and recognizing the state of workpieces in a bulk state due to deteriorated illumination conditions and similar colors with the background, making it challenging to accurately pick and convey multiple workpieces.

Innovation Solution

A target object recognition device comprising multiple calculation processing units, a state recognition unit, and a method determination unit that uses different techniques to calculate and recognize the attitude state of workpieces, and a manipulator with a robot arm controlled by a control unit to perform gripping operations based on the recognition results, integrated with an automatic guided vehicle for conveying.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If edge detection method is used to recognize workpiece position and attitude, then aligned workpieces can be picked accurately, but bulk state workpieces cannot be recognized due to deteriorated illumination and color similarity with background

Engineering Contradiction:

Improveworkpiece position and attitude recognition accuracyVSAvoidapplicability to different workpiece states

Core Design Contradiction:

Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The recognition system is segmented into multiple independent calculation processing units, each specialized for different workpiece states. The first calculation processing unit handles aligned workpieces using edge detection, while the second calculation processing unit handles bulk state workpieces using template matching and 3D information, allowing each segment to optimize for its specific function

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The recognition device is designed with multi-functionality to handle both aligned and bulk state workpieces. By integrating multiple calculation processing units with different recognition algorithms and selecting the appropriate unit based on workpiece state detection, the system achieves universal applicability across various workpiece arrangements

Inventive Principle:

Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If multiple calculation processing units with different techniques are used, then both aligned and bulk state workpieces can be recognized, but device complexity increases

Engineering Contradiction:

Improvecapability to handle aligned and bulk statesVSAvoidnumber of calculation processing units

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system dynamically selects which calculation processing unit to activate based on the detected workpiece state. The state recognition unit determines whether workpieces are in aligned or bulk state, and the method determination unit dynamically switches between different recognition methods, making the complexity adaptive rather than static

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system extracts and separates the different recognition functions into distinct calculation processing units. Instead of implementing a single complex unified recognizer, the patent extracts specialized recognition capabilities into separate units that can be independently selected and activated based on the specific workpiece state

Inventive Principle:

Principle #2Taking out (Extraction)

3Reliability

If edge detection is used for workpiece recognition, then illuminated and contrasting workpieces can be detected, but workpieces with same color as background or in poor illumination cannot be detected

Engineering Contradiction:

Improveedge detection reliabilityVSAvoidillumination conditions and color similarity

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent introduces template images as an intermediary for recognizing bulk state workpieces. Instead of directly detecting edges from the captured image which fails under poor illumination, the system uses template images stored in advance that represent typical workpiece appearances, allowing recognition to proceed even when direct edge detection is unreliable

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary action by storing template images of workpieces in advance before the actual recognition task. These pre-prepared templates capture the essential features of workpieces under various conditions, enabling the recognition system to compare against known patterns rather than relying solely on real-time image quality

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11964399B2Target object recognition device, manipulator, and mobile robot

Publication Date: 2024.04.23 OMRON CORP

AI summary

Provided is art capable of recognizing the states of a plurality of target objects arranged in a prescribed space region. This target object recognition device is provided with: a plurality of calculation processing units (21, 22) which each calculate the attitude state of a target object in a prescribed space region using a different technique; a state recognition unit (23) which recognizes the layout state of all of a plurality of target objects arranged in the space region; a method determination unit (24) which, in accordance with the result of the recognition by the state recognition unit (23), determines a method for using the results of the calculation performed by the calculation processing units (21, 22); and a target object recognition unit (25) which recognizes the attitude states of the target objects by means of the determined method for using the results of the calculation.