Method and system for detecting load-handling devices

Abstract

A computer-implemented method of identifying a load-handling device 850, 860 in a grid-based tracked 840 storage system 800 (and Fig 1) comprises obtaining an image of system 800 (and 410, Fig 4) by image sensors / camera 810, 820. The image is processed (420) with an object detection model trained to detect when a load-handling device 850, 860 has a unique similar pattern / identifier. The processing determines if the image includes a load-handling device 850, 860 (and 430 Fig 4) and outputs data based on the result (440). The object detection model comprises a convolutional neural network. An associated computer program, computer readable medium, data processing system, and system per se are also disclosed. Fig 8