Dynamic model identification of unmanned surface vehicles using deep learning network. (September 2018)

• Record Type: Journal Article
• Authors: Woo, Joohyun; Park, Jongyoung; Yu, Chanwoo; Kim, Nakwan
• Is Part Of: Applied ocean research. Volume 78(2018)
• Page Start: 123
• Page End: 133
• Publication Date: 2018-09
• Subjects: Unmanned surface vehicle (USV) -- System identification -- Deep learning -- Recurrent neural network (RNN) -- Long short-term memory (LSTM); Ocean engineering -- Periodicals; 620.416205
• DOI: 10.1016/j.apor.2018.06.011
• Languages: English
• ISSNs: 0141-1187

Abstract

Highlights: A deep learning–based dynamic model identification method is proposed to reduce estimation errors of USVs. To capture vehicle's dynamic behavior, a number of actuator test and free running test have been conducted. LSTM based recurrent neural network structure is used to grasp dynamic behavior of USV. Compared to the conventional model, proposed model reduced yaw rate error by 60.7% and sway velocity error by 27.9 %. Abstract: In this paper, a deep learning–based dynamic model identification method is proposed. The proposed method is designed to capture higher-order dynamic behaviors that result from the coupling of hydrodynamics and actuator dynamics. By adopting recent advancements in deep learning, our model addresses problems such as the regression problem in machine learning. Among various deep learning algorithms, long short-term memory (LSTM)–based recurrent neural network was used to deal with the hidden latent state of the USV dynamic model. The model validation was performed using free running test data of a USV. Analysis result shows that proposed model reduces surge speed prediction error by 76.9%, yaw rate prediction error by 60.7% and sway velocity prediction error by 27.9% over the conventional linear dynamic model.