AI Vision-Guided CMM Inspection for Mixed Workpiece Handling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional Coordinate Measuring Machines (CMMs) rely heavily on human operators for setup and data entry, limiting their ability to operate independently and making the inspection process less efficient, as they require manual handling and data input for workpiece identification and measurement.

Innovation Solution

A method and system that utilize artificial intelligence and camera systems to automatically identify and handle non-identical workpieces by analyzing images, retrieving digital product definitions, and controlling robots and inspection instruments to perform inspections, thereby reducing human intervention and error.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional CMMs are used with manual setup and data entry, then human operators can make decisions and prepare parts for inspection, but the inspection process becomes less efficient and requires continuous human intervention

Engineering Contradiction:
ImproveManual operation capabilityVSAvoidInspection process efficiency
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The system enables CMMs to automatically identify workpieces using vision systems, retrieve inspection parameters from digital product definitions, and perform measurements without human intervention. The automated workpiece identification and setup processes allow the system to serve itself, eliminating the need for manual data entry and preparation while maintaining operational capability.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

Manual mechanical operations are replaced with automated vision-based identification systems and computer-controlled measurement processes. The system substitutes human operators with automated algorithms that recognize workpiece features, retrieve inspection parameters, and control measurement operations, thereby improving efficiency while eliminating continuous human intervention.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If manual data entry and workpiece preparation are required, then flexibility in handling different workpiece types is maintained, but time is lost in setup and data input for each workpiece

Engineering Contradiction:
ImproveWorkpiece type flexibilityVSAvoidSetup and data input time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by automatically capturing workpiece images and identifying features before inspection begins. Digital product definitions are pre-configured with inspection parameters, allowing the system to rapidly retrieve and apply appropriate settings for different workpiece types without manual setup time, thereby maintaining versatility while eliminating setup delays.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The vision system creates digital copies of workpiece images and features, which are then used to automatically identify workpiece types and retrieve corresponding inspection parameters from digital product definitions. This copying approach eliminates the need for manual data entry while maintaining the ability to handle different workpiece types through digital pattern recognition.

Inventive Principle:
Principle #26Copying

3Productivity

If automated inspection systems are implemented, then productivity and efficiency are improved, but the system requires advanced capabilities in artificial intelligence and image analysis

Engineering Contradiction:
ImproveInspection throughputVSAvoidAI and vision system complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The vision system is designed with multi-functionality, serving as both a workpiece identification tool and a feature recognition system. The same image processing capabilities are used to identify workpiece types, locate features, and guide measurement operations, thereby achieving high productivity without requiring separate specialized systems for each function, thus managing complexity while maintaining automation.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS20230191634A1Multistep Visual Assistance for Automated Inspection
Publication Date: 2023.06.22 HEXAGON METROLOGY INC
  • US20230191634A1 patent drawing
  • US20230191634A1 patent drawing
  • US20230191634A1 patent drawing

AI summary

Illustrative embodiments provide a method by which artificial intelligence in combination with vision systems or cameras cooperate with a robot to automate a process for inspecting a workpiece. An illustrative method includes providing a set of cameras to image a set of workpieces that are randomly disposed in a storage area. A controller employing a neural network trained to identify workpieces then processes images from the set of cameras to identify each workpiece, and uses workpiece identity to customize the operation of an inspection system.