New AI-powered closed-loop tools can revolutionize manufacturing in 9 ways

A software company applies artificial intelligence in nine key ways to empower workers, improve performance and enhance productivity in closed-loop machine-vision-based applications.

 

Learning Objectives

  • Understand how artificial intelligence applied to machine vision in a closed-loop system can revolutionize manufacturing in nine ways.
  • Examine ways that artificial intelligence, machine vision, hardware and software, integrated in an easy to apply system, can empower workers, improve performance and enhance manufacturing productivity.

 

Machine learning insights

  • A new closed-loop machine-vision-based application helps manufacturing workers improve performance and enhance productivity.
  • An intuitive setup tool builds models in minutes instead of weeks or months.

Artificial intelligence (AI) helps manufacturers empower workers, improve performance and enhance productivity with a closed-loop machine-vision-based application for real-time worker training and quality control, according to Rapta, an applied AI software company based in Portland, Oregon. The company said its collaborative AI cameras and sensors can improve manufacturing yield with:

  • 90% lower rework costs
  • 98% faster inspection
  • 5-times faster training time
  • 6-month return on investment (ROI)
  • 1-day setup and use

The Rapta AI Supercoach and AI Supervisor systems are self-contained AI solutions that “set a new standard in real-time worker training and quality control, helping manufacturers solve today’s toughest challenges — without the need for a cloud connection,” the company said in an Oct. 17 statement.

Manufacturing applications for AI machine vision

AI-enabled machine-vision systems can resolve some of the most pressing manufacturing problems, Rapta said, including:

  • Day-one production training: Upskill workers to achieve production goals from day one.
  • Real-time quality control: Maintain consistent quality in manual and automated production.
  • Automated work instructions: Create and maintain video-based work instructions without manual intervention.
  • To exceed capabilities of traditional vision systems, Rapta said an AI-based machine vision system should excel in nine key areas.

Closed-loop work recognition

A closed-loop work recognition system virtually eliminates defects by actively training factory workers with video-based work instructions. This approach integrates AI-driven recognition of assembly tasks in real time, ensuring higher quality and efficiency across production lines.

Full traceability with photographic evidence

Full traceability for every step of every assembly captures video and photographic evidence to ensure comprehensive records for each product. A “digital traveler” provides transparency and compliance with strict quality standards.

Rapid AI vision model training

Unlike traditional systems that require weeks or months to build AI vision models, intuitive visual setup tool builds functional models in minutes. By using just a few examples (correct and incorrect), software generates AI training sets that are augmented for lighting, color and geometry, eliminating the need for thousands of manually captured images and engineering expertise.

No-code assembly builder

A no-code interface allows users to visually string together intricate manufacturing processes. Each step is represented as a visual tile, empowering users to add, delete, or rearrange processes with full version control, without coding or data science skills.

Full integration with digital tools

By integrating with connected tools such as torque drivers, multimeters and fluid dispensers, an AI system can ensure seamless validation of tool sequences and real-time monitoring. This integration maintains consistent quality and efficiency throughout the production process.

The Rapta’s AI Supercoach is said to be fast and easy to integrate into an existing manufacturing environment. Rapta said each system includes AI Supercoach Software, an AI computer (industrial PC, IPC), industrial Power over Ethernet (PoE) camera, mounting gantry, LED lighting, touch screen (industrial panel PC), power supplies and industrial PoE switch. Courtesy: Rapta
The Rapta’s AI Supercoach is said to be fast and easy to integrate into an existing manufacturing environment. Rapta said each system includes AI Supercoach Software, an AI computer (industrial PC, IPC), industrial Power over Ethernet (PoE) camera, mounting gantry, LED lighting, touch screen (industrial panel PC), power supplies and industrial PoE switch. Courtesy: Rapta

Human-like AI vision

An AI platform that mimics human vision makes it highly resilient to changes in lighting, color and physical orientation. The system leverages multiple specialized AI models, including those for reading text, ingesting tool data and understanding component orientation, achieving human-level accuracy in electro-mechanical assemblies.

Comprehensive revision control

Software streamlines the release management process with full revision control for video work instructions. New revisions require digital supervisor sign-off before updates are implemented, ensuring accountability and compliance with quality management systems. The system also allows for instant rollback or forward of approved releases, accommodating variant builds on the fly.

Mechanical scene memorization

Deep memorization capability allows vision system to adapt to mechanical float with human-like accuracy. This system can learn from one assembly in a standard orientation and maintain high accuracy even with rotations of up to 45º and positional floats of up to 20º.

Real-time continuous quality control

AI delivers real-time monitoring and step-locks workers when errors are detected, offering on-screen guidance with video playback until remediation is achieved. For post-assembly QA inspections, the system generates detailed reports with photographic evidence of issues that need correction.

Edited from information provided by Rapta by Mark T. Hoske, editor-in-chief, Control Engineering, WTWH Media, [email protected].

CONSIDER THIS

Are you getting rapid results from manufacturing AI and machine vision applications?

ONLINE

Control Engineering provides related articles about AI in manufacturing.

Control Engineering provides related articles about machine vision.

Written by

Mark T. Hoske

Mark Hoske has been Control Engineering editor/content manager since 1994 and in a leadership role since 1999, covering all major areas: control systems, networking and information systems, control equipment and energy, and system integration, everything that comprises or facilitates the control loop. He has been writing about technology since 1987, writing professionally since 1982, and has a Bachelor of Science in Journalism degree from UW-Madison.