报告题目: Modeling and Analysis of Motion Data from Dynamic Positioning Vessel for the Estimation of Environment

主 讲 人：程徐

报告时间：1月14日（周一）14:00

报告地点：屏峰校区计算机大楼B303

报告摘要:

The digital agenda is one of the pillars of the European strategy for growth, which proposes to increase Europe’s exploitation of the potential of information and communication technologies to foster innovation, economic growth, and progress. The strategy lists “ship intelligence” as one of the main areas through which to achieve growth. Ship intelligence has become a key aspect of making the maritime and offshore industries more innovative, efficient, and fit for future operations.

As the complex marine operations have been moving towards the ultra-deep sea, sensing the environment becomes more and more important. Therefore, developing a real-time and reliable model to estimate the environment is significant to aid the decision making for the autonomous ship. We proposed a novel deep neural network-based model, which mainly consists of three components: an LSTM recurrent neural network to capture the long dependency in the ship motion data; a convolutional neural network (CNN) to extract time-invariant features; and a Fast Fourier Transform (FFT) block to extract frequency features. A feature fusion layer is designed to learn the degree affected by each component. The proposed model is evaluated in the benchmark datasets and ship motion dataset, and the experiments on those datasets demonstrate its feasibility. The investigation on real-time testing verifies the practicality of the proposed model.

报告人简介

Xu Cheng received the M.S. degree in computer science and technology from Zhejiang University of Technology, Hangzhou, China, in 2015. He is currently working toward the Ph.D. degree at the Department of Ocean Operations and Civil Engineering, Mechatronics Laboratory, , Aalesund, Norway.

His current research interests include sensitivity analysis, neural network, big data and ship motion modeling.