HandCOMM: AI Solution for Crane & Workers’ Safety

The construction industry heavily relies on mobile cranes for the off-site modular construction, emphasizing safety in operations. Despite standardized hand signals, inadequate and interrupted communications between mobile crane signalmen on the ground and crane operators pose significant risks of accidents, especially during blind lifts or lifts in severe weather. This research project aims to develop a prototype addressing communication problems between crane operators and signalmen in mobile crane operations by designing robust embedded systems in the glove and helmet and a live data receiver module. These embedded systems, equipped with flex sensors, Inertial Measurement Units (IMUs), cameras, and data transfer capabilities, enhance the recognition of hand gestures and facilitate real-time data transmission between modules. Two techniques, sensors-based and computer vision (CV) based approaches, are used to recognize hand gestures, which allow cross-validated and precise identification of hand signals. The crane operator’s cabin’s data receiver incorporates auditory and visual modalities to convey the information. The application of the prototype is expected to improve on-site communication efficiency and safety during blind lifts of crane operations. The anticipated outcomes have the potential to revolutionize communication protocols in the construction industry, ensuring safer and more efficient practices.

Faculty Supervisor:

Xinming Li;Jie Han

Student:

Partner:

SensiImage Technologies Ltd.

Discipline:

Engineering

Sector:

Construction and infrastructure

University:

University of Alberta

Program: