Hierarchical neural network-based hydrological perception model for underwater glider

The underwater glider is a robust ocean observation platform with low power consumption to achieve detailed ocean data. Hydrological information plays a key role in the ocean economy and environment evolution. It is often desirable to detect the hydrological and track their variation in the ocean observation task. This paper proposes a hierarchical neural network-based hierarchical perception model for the underwater glider to perceive the hydrological information in the ocean environment. First, a one-dimensional convolutional neural network is designed for detecting the thermocline layer. Then, the identified deep layer is predicted by a long short-term memory network. Furthermore, finite element analysis is conducted to explore the buoyancy loss of the underwater glider in the ocean. Finally, extensive experiments are conducted to demonstrate the effectiveness and accuracy of the proposed model.