ABB launches Fashion Inductor module with AI Robot Trainer to cut integration workload

ABB Robotics on Oct 15 introduced its first end-to-end, vision-based industrial Physical AI solution, integrating AI Robot Trainer into the Fashion Inductor module for automated sorting of apparel and accessories. The company said the combination can reduce system integrators' engineering and commissioning workload by up to 75%.

Fashion Inductor is designed for the high variability of apparel logistics, where items, packaging and operating conditions constantly change. According to ABB, the module can sort up to 1,300 items per hour per unit, with a target picking accuracy above 99.5%, supported by AI-driven path planning, motion precision and dedicated vision technology. The AI Robot Trainer provides pre-trained garment-specific vision models, application tools and rapid retraining capabilities, allowing integrators to deploy and continuously optimize without a dedicated AI team.

The accompanying digital twin is an exact replica of the module software, linking the Application Controller Platform, automatic path planning, AI Robot Trainer and RobotStudio to speed commissioning and support third-party components such as grippers. AI Robot Trainer's retraining follows a closed loop: monitoring robot performance, identifying anomalies, generating new training data and deploying improvements back to the physical robot. RobotStudio handles virtual design, simulation and verification, while the OmniCore controller acts as the execution layer on site.

ABB said Fashion Inductor is the first step in its vision-based Physical AI expansion, with the same AI Robot Trainer approach to be extended to more vision applications through the first quarter of 2027, aiming at more autonomous and versatile robots.