GPS assisted voice signalling for safe crane operation - usability study recording

Monash University research to improve safety in crane operations. Recording from a usability study. Abstract On construction sites, the signaller directs the motions of the crane and load with hand, whistle or voice signals. The standardised signals are vague and cumbersome, and many crane accidents have been attributed to failures of communication between the signaller and crane operator. This research proposes an extension to the current signalling system to support the current practice while increasing safety. The signaller marks a location with a handheld GPS device, and this is automatically annotated on a screen in the crane operator's cabin. The signaller then uses the standardised voice signals, while optionally referring to the marked location to specify how far the crane should travel. The solution has been informed by analysis of current practices, regulations, standards, causal factors of crane related accidents, and the maturity of available technologies. The solution significantly improves task performance, reduces mental demand, and improves congruence of task-related knowledge within the team. Chapters 0:00 Introduction 1:00 Slides (training) 7:48 Video 10:19 Interaction 19:26 Lift B1 (extended signalling) 22:47 Lift B2 25:00 Lift B3 27:10 Lift B4 30:03 Lift A1 (current signalling) 32:28 Lift A2 34:55 Lift A3 38:28 Lift A4 Ethics Declaration This research was undertaken with approval from the Monash University Human Research Ethics Committee (MUHREC) - project ID 47938. Acknowledgments The authors would like to thank the Monash Drone Discovery Platform for providing the mesh model of the construction site used for creating the experiment scenes. The 3D model `Warsaw High-Stored RCC Apartment Building' by Andrey Potiekha (Sketchfab user Imeant) from https://skfb.ly/prOAw was used, as modified by us. Technology Game engine: Unity 3D VR headset: Meta Quest 3 Controllers: 2x Logitech Extreme 3D Pro Joystick Audio editing: NCH WavePad Video editing: Shotcut Subtitles: Aegisub Subtitle hardcoding: FFmpeg