Accuracy of a ceiling vision localization system on an overhead crane

Lifting heavy cargo, parts or equipment is a critical part of any assembly line of a factory or a warehouse, and thus efficient use of cranes can make a world of difference. Sophisticated and intelligent crane positioning systems on the other hand can benefit the safety and overall usability of the device. With the latest camera technology, remote control and even automation of the crane has been made possible.
The goal of this study was to create and evaluate the accuracy of a positioning system for an overhead crane based on the stereo image captured from the ceiling above. The technology used to achieve this was Visual SLAM (Simultaneous Localization And Mapping). One of the main objectives was to construct an integrable alternative system for the crane’s own positioning system, which measures distance of the crane’s bridge and trolley from the walls using a laser. The study produced a dataset comparing the existing laser-based positioning system and the camera-based system.