Bio-inspired underwater robotic vehicle for marine exploration and AI-powered fish detection

This study presents the design and simulation of a novel bio-inspired underwater robotic vehicle (ROV) that emulates whale locomotion for agile marine exploration and real-time fish detection. Drawing inspiration from cetacean biomechanics, the proposed vehicle incorporates oscillating flukes, tubercled flippers, and a flexible dorsal ridge to replicate full-body aquatic propulsion. The structural design was developed using SolidWorks® and validated via six-degree-of-freedom (6-DOF) kinematic modeling. Computational fluid dynamics (CFD) simulations revealed a favorable thrust-to-drag ratio and reduced flow separation due to tubercle integration, with up to 22% drag reduction compared to baseline flipper geometries. To complement the mechanical platform, a novel artificial intelligence-based perception module called Ghost-YOLV12 is proposed, which is an enhanced version of the YOLOv12 deep learning model. Trained on the DeepFish dataset, the proposed model achieved a mean average precision (mAP50) of 97.8 and demonstrated robust performance under occlusion, turbidity, and low-light conditions. All evaluations were conducted in simulation environments, with hydrodynamic testing performed through CFD and fish detection validated through annotated datasets. While no physical prototype has been deployed yet, the design is fully scalable and structured for real-world fabrication. The manuscript discusses design limitations, deployment considerations, and future directions, including real-environment testing, embedded AI optimization, and energy-efficiency benchmarking. This work contributes a comprehensive bio-AI robotic framework that integrates biologically accurate locomotion with intelligent marine sensing, offering potential for ocean monitoring, fisheries research, and non-invasive ecological surveys. The AI implementation code is available online https://github.com/Aliweka2020/Ghost-YOLOv12