ROADMAP Robot-Aided Diver Navigation in Mapped Environments

Abstract

Technical dives are challenging as they often involve unhospitable sites with limited or no visibility, dangerous or cluttered environments and are under a constant influence of external disturbances. Diver augmented vision displays can play a significant role when it comes to assisting divers to fully understand tasks that need to be accomplished in these situations. In order for such a system to function properly, precise diver navigation and localization relative to the object of interest needs to be ensured. Further, to improve safety, integrating physiological instrumentation in the divers suit could enhance such a system by providing a picture of their physiology in real-time. The ROADMAP project builds upon recent research accomplishments in the area of human-robot interaction underwater, by ensuring precise diver navigation through the use of underwater robots that navigate using precise navigation sensor set and previously obtained maps of the underwater area of interest.Themain objectives of the project are to develop telemetry transmission protocols for wired and reliable wireless data exchange from the autonomous underwater vehicle BUDDY to the Diver Augmented Vision Display (DAVD) for the purpose of enhancing diver navigation; develop algorithms for precise diver navigation during free SCUBA diving by exploiting multisensory measurements onboard the BUDDY vehicle while tracking the diver; develop algorithms for precise diver localization in previously mapped environments by exploiting BUDDY vehicle map-based localization and sonar-based diver perception; to evaluate the feasibility of the proposed robot-aided diver navigation solutions in a real environment. Kinematic and physiological data obtained from wearable instrumentation in the divers suit will be used to infer diver physiological state; likelihood of fatigue and to infer exactly what the diver is doing in real-time. This can then be relayed to an VR monitor displaying a virtual workspace (e.g. wreck, reef, etc.) with an active diver avatar with realistic orientation of the body and activity in the workspace.

Document Details

Document Type
DoD Grant Award
Publication Date
May 05, 2021
Source ID
N000142112274

Entities

People

  • Nikola Mišković

Organizations

  • Faculty of Electrical Engineering and Computing, University of Zagreb
  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Materials Science
  • Sensor Fusion and Tracking Systems.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy