Towards Explainable Deep Learning for Ship Trajectory Prediction in Inland Waterways
Authors: Tom Legel, Dirk Söffker, Roland Schätzle, Kathrin Donandt
Organizations: Department Hydraulic Engineering in Inland Areas, Federal Waterways Engineering and Research Institute, Germany · Chair of Dynamics and Control, University of Duisburg-Essen, Germany · Department of Technology (Computer Science), Baden-Wuerttemberg Cooperative State University, Germany · Institute for Sustainable and Autonomous Maritime Systems, University of Duisburg-Essen, Germany
Accurate predictions of ship trajectories in crowded environments are essential to ensure safety in inland waterways traffic. Recent advances in deep learning promise increased accuracy even for complex scenarios. While the challenge of ship-to-ship awareness is being addressed with growing success, the explainability of these models is often overlooked, potentially obscuring an inaccurate logic and undermining the confidence in their reliability. This study examines an LSTM-based vessel trajectory prediction model by incorporating trained ship domain parameters that provide insight into the attention-based fusion of the interacting vessels' hidden states. This approach has previously been explored in the field of maritime shipping, yet the variety and complexity of encounters in inland waterways allow for a more profound analysis of the model's interpretability. The prediction performance of the proposed model variants are evaluated using standard displacement error statistics. Additionally, the plausibility of the generated ship domain values is analyzed. With an final displacement error of around 40 meters in a 5-minute prediction horizon, the model performs comparably to similar studies. Though the ship-to-ship attention architecture enhances prediction accuracy, the weights assigned to vessels in encounters using the learnt ship domain values deviate from the expectation. The observed accuracy improvements are thus not entirely driven by a causal relationship between a predicted trajectory and the trajectories of nearby ships. This finding underscores the model's explanatory capabilities through its intrinsically interpretable design. Future work will focus on utilizing the architecture for counterfactual analysis and on the incorporation of more sophisticated attention mechanisms.
Figures & tables
Figure 1: Illustration of positional features wkm as kit , offset from the fairway axis as fit , and their changes with respect to the previous time step Δkit and Δfit .
Figure 2: Illustration of ship-to-ship relations for opposing ships ( Θ=−1 ). The change in longitudinal distance Φijt is given by Δijt−Δijt−1 .
Figure 3: Model variants
Relative
Distance
Lateral
direction Θ
change Φ
distance Γ
(wkm/min)
(m)
-1 (opposing)
<−0.2
0−10
1 (aligned)
−0.2 to −0.05
10−20
0 (stationary)
−0.05 to +0.05
20−40
–
>0.05
>40
Table 1: Discretisation of ship-to-ship relation values.
Training
Validation
Test
Situations
15,920
2,079
2,055
Total
159,673
21,178
20,758
trajectories
Table 2: Sizes of datasets in number of trajectories and situations.
Figure 4: FDEt per model and time steps. The error values are linearly scaled from 0 to 50 and log10 -scaled beyond that range. Outliers are shown as diamonds.
Model
E-D
EA-DA
E-DA
E-DDA
Mean
45.1
41.9
38.4
40.9
Median
31.2
28.8
25.1
27.3
σ
64.7
61.9
60.1
61.0
Table 3: Statistic values of FDE5 .
Figure 5: Ship domain values for vessels that are moving in opposite directions. The red dashed line indicates the initial ship domain value.
Accurate long-term trajectory prediction for maritime vessels is essential for safety and logistical efficiency. While deep learning models, particularly Transformers, have shown promise in processing Automatic Identification System (AIS) data, they often struggle with the quadratic computational complexity of self-attention and the loss of accuracy over extended forecasting horizons. This study proposes TempTPI, a novel prediction framework that integrates an Informer-based encoder with a multi-channel temporal encoding mechanism. The Informer architecture leverages a ProbSparse self-attention mechanism to reduce computational overhead and focus on the most significant dependencies, while the temporal encoder utilizes Fourier-like frequency expansions to capture cyclic patterns (hourly, daily, and seasonal) in vessel behavior. We evaluate our model against the state-of-the-art TPTrans architecture using AIS data from Danish waters. Experimental results demonstrate that TempTPI consistently outperforms existing methods across prediction windows of 1 to 5 hours. Notably, at a 5-hour horizon, the proposed model achieves a 55% improvement in Mean Squared Error (MSE), offering a robust solution for long-range maritime situational awareness.
Kevin Ferneding, Veronika Lietavcova, Aleksandra M. Blachowiak +1
Technical University of Denmark Anker Engelunds Vej 101 2800 Kongens Lyngby, Denmark
Recent advancements in reinforcement fine-tuning have significantly improved the reasoning ability of large language models (LLMs). In particular, methods such as group relative policy optimization (GRPO) have demonstrated strong capabilities across various fields. However, applying LLMs to ship trajectory prediction remains largely unexplored. In this paper, we propose ShipTraj-R1, a novel LLM-based framework that reformulates ship trajectory prediction as a text-to-text generation problem. (1) We design a dynamic prompt containing trajectory information about conflicting ships to guide the model to achieve adaptive chain-of-thought (CoT) reasoning. (2) We introduce a comprehensive rule-based reward mechanism to incentivize the reasoning format and prediction accuracy of the model. (3) Our ShipTraj-R1 is reinforced through the GRPO mechanism guided by domain-specific prompts and rewards, and utilizes the Qwen3 as the model backbone. Extensive experimental results on two complex and real-world maritime datasets show that the proposed ShipTraj-R1 achieves the least error compared with state-of-the-art deep learning and LLM-based baselines.
Yang Zhan, Yunhao Li, Zhang Chao +2
School of Artificial Intelligence, Optics, and Electronics (iOPEN), Northwestern Polytechnical University, Xi’an 710072, China · Department of Logistics and Maritime Studies, The Hong Kong Polytechnic University, Hong Kong 999077, China · State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China
To address the challenges of behavioral multimodality, limited semantic utilization, and long-term error accumulation in vessel trajectory prediction, this paper proposes M3-Former, a multimodal trajectory prediction framework enhanced by large language models (LLMs). The proposed framework incorporates vessel static attributes and navigational intent as semantic priors for long-term trajectory modeling. Specifically, a unified multimodal representation space is constructed, in which static semantic information is encoded by a pre-trained LLM and aligned with dynamic trajectory features through self-attention. To jointly capture global route planning and local motion variations, a dual-granularity Mixture-of-Experts (MoE) architecture is introduced, where sequence-level experts model global navigation trends and token-level experts refine fine-grained maneuvering behaviors. In addition, a Steering-Weighted Cross-Entropy loss is designed to alleviate the long-tail distribution of sparse turning samples and improve prediction accuracy in critical maneuvering scenarios. Experiments on a real-world Danish AIS dataset demonstrate that M\textsuperscript{3}-Former consistently outperforms state-of-the-art baselines across prediction horizons from 1 to 4 hours. In the 4-hour prediction task, the proposed method reduces Average Displacement Error (ADE) and Final Displacement Error (FDE) by 4.4% and 5.1%, respectively, compared with the strongest baseline. Qualitative and ablation analyses further verify that semantic fusion effectively reduces long-term trajectory drift, while the dual-granularity MoE improves robustness in complex waterways and route-branching scenarios. The proposed framework establishes a semantic-guided hierarchical prediction paradigm, in which high-level navigational intent and local motion dynamics are jointly modeled for robust long-term vessel trajectory forecasting.
Wenzhe Jin, Haina Tang
School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China