21 - 25 April 2024
National Harbor, Maryland, US
Conference 13061 > Paper 13061-4
Paper 13061-4

Eye on the back: augmented visuals for improved ROV teleoperation in deep water surveillance and inspection

23 April 2024 • 1:50 PM - 2:10 PM EDT | Potomac 5

Abstract

Underwater ROVs (Remotely Operated Vehicles) play a crucial role in subsea inspection, remote surveillance, and deep-water explorations. Typically, a surface operator controls the ROV based on its real-time camera data, which is first-person visual feedback. However, underwater ROVs’ onboard camera feed only offers a low-resolution and often noisy egocentric view - that is not very informative in deep water and adverse visual conditions. To address this, we introduce the “Eye On the Back (EOB)” technology to provide a third-person view for improved underwater ROV teleoperation. Integrating EOB views to teleoperation consoles facilitates interactive features with augmented visuals for the teleoperator as well as for enabling semi-autonomous behavior such as next-best-view planner and active ROV localization. We conduct a series of field experiments to validate this technology for remote ROV teleoperation in underwater cave exploration and subsea structure inspection tasks. We are currently developing an end-to-end portable solution that can be integrated into existing ROV platforms for general-purpose subsea telerobotics applications.

Presenter

Md Jahidul Islam
Univ. of Florida (United States)
Underwater ROVs play a crucial role in subsea inspection, remote surveillance, and deep-water explorations. Typically, a surface operator controls the ROV based on its real-time camera data, which is first-person visual feedback. However, underwater ROVs' onboard camera feed only offers a low-resolution and often noisy egocentric view - that is not very informative in deep water and adverse visual conditions. To address this, we introduce the 'Eye On the Back (EOB)' technology to provide a third-person view for improved underwater ROV teleoperation. We conduct a series of field experiments to validate this technology for remote ROV teleoperation in underwater cave exploration and subsea structure inspection tasks. We are currently developing an end-to-end portable solution that can be integrated into existing ROV platforms for general-purpose subsea telerobotics applications.
Application tracks: AI/ML , Sustainability
Presenter/Author
Md Jahidul Islam
Univ. of Florida (United States)