Predicting trajectories of coastal area vessels with a lightweight Slice-Diff self attention

• Published in: Complex & intelligent systems Vol. 11; no. 5; pp. 239 - 17
• Main Authors: Zhang, Jinxu, Liu, Jin, Zhang, Xiliang, Wei, Lai, Wu, Zhongdai, Wang, Junxiang
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.05.2025; Springer Nature B.V; Springer
• Subjects: Automatic identification system; Coastal collision avoidance; Coastal zone; Coasts; Complexity; Computational Intelligence; Data Structures and Information Theory; Engineering; Identification systems; Intelligent systems; Irregular trajectory feature extraction; Kalman filters; Maritime areas; Neural networks; Original Article; Recurrent neural networks; Structured data; Trajectory slice and difference; Vessel trajectory prediction; Automatic identification system; Trajectory slice and difference; Irregular trajectory feature extraction; Coastal collision avoidance; Vessel trajectory prediction
• ISSN: 2199-4536; 2198-6053
• DOI: 10.1007/s40747-025-01877-x

Accurate prediction of vessel trajectories in coastal areas poses a significant challenge due to the large number of irregular trajectories. Existing trajectory prediction studies predominantly employ recurrent neural network (RNN) and Transformer-based methods. However, the former often encounter challenges such as gradient vanishing or exploding, and the latter tend to focus on global temporal dependencies, making it difficult to capture local irregular trajectory features in coastal maritime areas. Recently, graph-based methods have also been used to predict trajectories, however processing graph-structured data introduces significant increase in computation. In responding to these, this paper proposes a framework based on a novel lightweight Slice-Diff self attention, which consists of several key components. Firstly, the trajectory slice difference encoder (TSDE) utilizes slice embedding (SE) to enrich the cross dimensional dependencies contained in the input sequence, and then combines Slice-Diff self attention (SDSA) and fine-grained convolution (FGC) to comprehensively capture sequence-specific positional and directional information. Additionally, an auxiliary model, stepping bidirectional long short-term memory (S-BiLSTM) is developed to capture global temporal dependencies within the whole sequence. Finally, the fine-grained trajectory features obtained from TSDE and the global temporal dependencies compensated by S-BiLSTM are combined through the fully connected layer to predict coastal vessel trajectories. Extensive experimental results on three real-world automatic identification system (AIS) datasets demonstrate the effectiveness of proposed framework against other baselines.