Underwater Search and Rescue Robot Based on Convolutional Neural Network

Abstract

In this paper, a low-cost underwater search and rescue robot is developed, which aims to improve the efficiency of underwater rescue and to solve the problems of high cost and difficult deployment of the mainstream equipment currently on the market. The device uploads search and rescue data to the management and dispatch platform through the 4G module. The search path and area of each robot are set by the management scheduling platform. Use deep learning to process the data collected by the camera to complete the high-precision detection of drowning targets.