Intelligent Assisted Navigation for Ships Using Edge Computing and Neural Networks

In order to meet the requirements of high real-time and stability in the process of ship navigation, this paper applies an intelligent auxiliary means based on edge computing and convolutional neural network (CNN). The edge computing platform is deployed on the ship, and sensors and cameras are used to collect surrounding data. The convolutional neural network models are trained to accurately recognize and classify different types of obstacles, and real-time environmental perception and analysis are performed. Intelligent driving assistance decisions such as automatic obstacle avoidance, path planning, and speed control are provided based on the analysis results. The research results indicate that the average recovery time of the system is between 5.8 minutes and 29.5 minutes, with an accuracy of 0.97 in obstacle recognition tasks. The method can achieve intelligent assistance for ship navigation with high real-time performance and stability.

I. Introduction

Navigation of ships at sea is a high-risk activity, which is particularly important for the safety of crew members. In marine environments, the driving process typically faces multiple challenges. During adverse weather conditions, limited visibility increases driving risks due to heavy fog, heavy rain, and waves. The positions of various obstacles such as other ships, buoys, and reefs are not fixed, which requires crew members to identify and respond in a timely manner. The application of intelligent assisted driving technology has become an inevitable choice for improving ship driving safety. Edge computing and neural network technology can help to perceive and analyze the marine environment in real time, provide more timely and accurate ship navigation and driving assistance, and reduce accident risk.