Low-Cost, Long-Endurance Cooperative Navigation Based on “Light” Marine Equipment in Deep Sea

Low-cost, long-endurance navigation is a crucial universal supporting technology for long-range underwater vehicles while entering, exploring, and developing the deep sea. To overcome the limitations of existing navigation systems (e.g., high cost and unsuitability for long-endurance operation), this study proposes a cooperative navigation system based on “light” marine equipment such as wave glider, underwater acoustic communication modem (UACM), and micro-electro-mechanical system (MEMS) sensor. The focus was to resolve the crucial technical problems faced by the above mentioned “light” marine equipment during deep integration. First, the wave glider path planning based on deep reinforcement learning is studied to ensure continuous and reliable information interaction between the wave glider and underwater vehicle. Second, underwater acoustic communication one-way ranging based on a sound velocity profile (SVP) coupling model and a constant gradient SVP fast algorithm is studied to achieve fast and accurate estimation of the horizontal range between the wave glider and underwater vehicle. Finally, UACM-MEMS sensor cooperative navigation based on decision feedback-denoising auto-encoding and adaptive Kalman filter is studied to ensure the accuracy of navigation. Through deep integration of each “light” marine equipment, low-cost, long-endurance cooperative navigation can be realized in the deep sea.