Dynamic Detecting Based Trajectory Planning for AUV to Collect Data from Underwater Sensors

Abstract

Ocean big data is becoming a future trend of the Internet of Underwater Things (IoUT). Underwater wireless sensor networks (UWSNs) technique is a promising method to realize ocean big data. However, the limited energy and the low location accuracy of sensor nodes make the data collection of UWSNs difficult. To reduce the energy consumption of sensor nodes, we consider an autonomous underwater vehicle (AUV) based data collection, where the AUV moves close to the sensors to collect data using short-range high-rate communications. Particularly, we propose a grouping-based dynamic trajectory planning (GDTP) for the AUV. GDTP does not require the position information of sensor nodes. It dynamically detects the existence and directions of sensor nodes, based on which the detected sensor nodes are grouped by a proposed common communication area model. The cruising direction of the AUV is dynamically determined with the maximum expected payoff that considers the data collection and energy consumption in inaccurate detection. Simulation results show that the proposed GDTP collects more data packets with less energy compared to the existing schemes.