Hanmin Jiang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Hanmin Jiang has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Jiangjin Yin
College of Informatics, Huazhong Agricultural University, Key Laboratory of Smart Farming for Agricultural Animals, Hubei Engineering Technology Research Center of Agricultural Big Data, and Engineering Research Center of Intelligent Technology for Agriculture, Wuhan, China
1 shared publication - Qianzhen Liu
College of Computer Science and Electronic Engineering, Hunan University, China
1 shared publication - Ye Xiao
College of Computer Science and Electronic Engineering, Hunan University, China
1 shared publication - Yi Xiao
School of Electrical Engineering, North China University of Water Resources and Electric Power, China
1 shared publication - Yupeng Hu
College of Computer Science and Electronic Engineering, Hunan University, Changsha, China
1 shared publication
Latest Publications
- Open Access
- 24 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Attention Mechanism-based Forward and Backward Data-Driven Method for Ship Trajectory Prediction
As global seaborne trade increases, ship trajectory prediction becomes more practical and urgent. However, accurate prediction of its next move remains challenging due to the variability and uncertainty associated with ships. Thus, we propose an attention mechanism-based forward and backward data-driven method (termed Att-FB). The method employs a forward gated recurrent unit (GRU) with an attention mechanism (GRU-Attention) neural network block, a backward bidirectional GRU (BiGRU) with an attention mechanism (BiGRU-Attention) neural network block, and fusion network block composed of multi-layer perceptrons (MLP). The attention mechanism is employed in Att-FB to improve the feature learning ability of ship behavior data. We also incorporate the backward historical trajectory segment of the target prediction section into the prediction method for enhancing learning ability. Enough experiments were conducted on the American automatic identification system (AIS) dataset to verify the superiority of our Att-FB method. The Att-FB method demonstrates an 85.42% improvement on average in prediction accuracy compared to the baseline methods. Moreover, we also study the prediction step size and sampling time optimal hyperparameters. The results of our study show that the Att-FB method has good prediction performance, providing valuable guidance for marine transport applications.
- 24 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessAttention Mechanism-based Forward and Backward Data-Driven Method for Ship Trajectory Prediction
As global seaborne trade increases, ship trajectory prediction becomes more practical and urgent. However, accurate prediction of its next move remains challenging due to the variability and uncertainty associated with ships. Thus, we propose an attention mechanism-based forward and backward data-driven method (termed Att-FB). The method employs a forward gated recurrent unit (GRU) with an attention mechanism (GRU-Attention) neural network block, a backward bidirectional GRU (BiGRU) with an attention mechanism (BiGRU-Attention) neural network block, and fusion network block composed of multi-layer perceptrons (MLP). The attention mechanism is employed in Att-FB to improve the feature learning ability of ship behavior data. We also incorporate the backward historical trajectory segment of the target prediction section into the prediction method for enhancing learning ability. Enough experiments were conducted on the American automatic identification system (AIS) dataset to verify the superiority of our Att-FB method. The Att-FB method demonstrates an 85.42% improvement on average in prediction accuracy compared to the baseline methods. Moreover, we also study the prediction step size and sampling time optimal hyperparameters. The results of our study show that the Att-FB method has good prediction performance, providing valuable guidance for marine transport applications.