Underwater multi-robot autonomy with visual language models, multimodal AI, 3D mapping, coral and infrastructure monitoring, and adaptive power-aware operation.
MARVEL integrates underwater multi-robot autonomous systems with visual language models and multimodal AI to enable wide-area monitoring in underwater environments. Collaboration among autonomous underwater vehicles enables assessment of water quality and environmental health, producing 3D mapping of physical and chemical environmental parameters.
Advanced computer vision models allow online and offline observation of marine life, particularly coral reefs, and man-made structures like aquaculture farms, oil and gas facilities, and ports. MARVEL develops unified multi-modal 3D mappings for intelligent vision-language tasks such as Visual Question Answering, image captioning, and Text-to-Image search.
Figures
