Become a member
Take advantage of exclusive member benefits, world class events, networking and specialist support
17 November 2026
ABB says robotic AI vision can be continuously improved by applying its new training module, without relying on expert help to build models.
By expanding its OmniCore EyeMotion system with AI Robot Trainer, ABB Robotics says it is enabling customers to train, deploy, and continuously improve AI-powered robot vision applications faster, and with no requirement for external specialists to construct models from scratch.
Designed for system integrators and automation partners, OmniCore EyeMotion powered by AI Robot Trainer simplifies the deployment and scaling of AI-powered vision applications, says the company, accelerating commissioning and adaptation to production changes.
Converting real shop floor images into deployable AI vision data, the tools connect image acquisition, AI model training, deployment and continuous improvement in a single industrial workflow, it explains. The system is able to convert perception directly to robot motion through automatic collision-free path planning.
According to president Marc Segura, physical AI fundamentally changes what robots can do, where they can operate and the value they can create.
“Robotic AI vision is the vital bridge between robot perception and physical actions, enabling robots to see, understand, and act in real time,” he says. “Customers need a complete workflow to create, deploy and continuously improve it. OmniCore EyeMotion and AI Robot Trainer work together, in a continuous loop of data generation, AI model training, deployment and optimisation.”
ABB claims the integrated system reduces commissioning time by up to 90% and cycle time by up to 50% when combined with automatic collision-free path planning. A simple guided workflow and AI chatbot lead users through every step of the process, helping system integrators handle changing SKUs, lighting, layouts and object variations.
01908 350300
www.abb.com