Underwater smart glasses: A visual-tactile fusion hazard detection system

Area of Science:

• Robotics and Automation
• Marine Technology
• Sensory Systems

Background:

• Underwater operations face significant risks from marine hazards and collisions.
• Current safety measures for divers are limited in their ability to provide comprehensive, real-time hazard detection.

Purpose of the Study:

• To develop and validate underwater smart glasses (USGs) integrating visual and tactile sensing for all-around underwater hazard detection.
• To enhance operational safety for divers by providing real-time alerts for marine animal attacks and collisions.

Main Methods:

• Developed a vision module using an AI-enabled optical sensing system and a novel underwater target detection algorithm (YOLO-UH).
• Designed a tactile module utilizing kirigami tribo-materials (KTMs) for mechanical sensing and collision detection.
• Employed visual-tactile fusion to combine data from both sensing modules for comprehensive hazard awareness.

Main Results:

• The YOLO-UH algorithm effectively detected toxic marine organisms within the diver's visual field.
• The KTM-based tactile module demonstrated sensitive detection and warning capabilities for collisions outside the visual field.
• Validated the performance of both modules through numerical simulations, laboratory tests, and real-world experiments.

Conclusions:

• The proposed USG system, with its visual-tactile fusion, provides near-far, all-around hazard detection capabilities.
• This technology significantly reduces the risks associated with underwater activities for divers.
• The developed USG represents a novel approach to improving underwater operational safety.