An underwater wake sensing method based on Micro Thermoelectric Generator for enhancing recognition accuracy

Abstract

The accidental entry and poaching of foreign boats and autonomous underwater vehicles (AUVs) have a detrimental impact on the ecological environment of marine ranching. Currently, there is a need for a monitoring method specifically designed to monitor underwater moving objects (UMOs), such as foreign boats and AUVs, in real-time within marine ranching. Therefore, this paper creatively proposes a method for monitoring UMOs based on thermal tactile of Micro Thermoelectric Generator (MTEG). A thermal tactile perceptual model for propeller wake identification is established and the wake sensing sensor (WSS) is designed. Furthermore, the performance verification experiment of wake sensing and the target classification experiment based on machine learning are carried out to prove the feasibility of the method. The experimental results demonstrate that the propeller wakes are identified with an accuracy of more than 90 % within 30 s in the monitoring terminal. Besides, the method can accurately distinguish between propeller wakes and fish wakes, thereby avoiding false alarms. The reported underwater wake sensing method is a new solution to real-time monitoring and safety early warning of marine ranching.