Perception and Autonomous Navigation Framework for an Underwater Micro-Bionic Snail Robot

The exploration of the underwater environment has driven research into bionic robotic systems, particularly the development of efficient and compact underwater robots capable of performing complex tasks. This study proposes an advanced framework for the perception and autonomous navigation of underwater micro-bionic water snail robots in accordance with the current bio-mechanistic motion model. The proposed model combines a hybrid neural control system with reinforcement learning, with the objective of enhancing the adaptive navigation of the robot in a dynamic underwater environment. In particular, the robot perception system employs a multi-modal sensor fusion methodology to integrate visual, fluid dynamics, and acoustic signals, thereby ensuring the robust perception of the surrounding environment. Additionally, an innovative approach to path planning has been developed, utilizing a deep neural network as its foundation, aimed to enhance obstacle avoidance and trajectory optimization. Furthermore, the locomotion system mimics the peristaltic motion of a water snail, thereby enhancing energy efficiency and mobility in confined environments. This study presents two novel concepts: real-time feedback mechanisms for adaptive behavior and a hierarchical decision-making framework balancing immediate actions with long-term objectives. The experimental results indicate that the proposed model showcases considerable strengths in navigation precision, adaptability, and energy efficiency.