Unfolding AIS transmission behavior for vessel movement modeling on noisy data leveraging machine learning

TL;DR

The paper tackles forecasting AIS message content for multiple vessels under irregular transmission timing by framing AIS messages as events $\vec{v}=\langle \rho,\omega,\psi,\epsilon,\mu\rangle$ on trajectories $\tau_i$ and employing a multi-trajectory, multi-variable forecasting approach. It formalizes the problem using window length $w$ and horizon $s$, with $\Delta \mathrm{T}$ as a continuous-time feature, and models the mapping $h:\mathbb{R}^{|\bar{x}|}\to\mathbb{R}^{|\bar{y}|}$ across trajectories. The methodology combines windowed sampling with a two-block Conv1D-LSTM architecture and an autoregressive path to forecast the full AIS message content over horizons, guided by a robust optimization regime using AdamW and the Hyperbolic Tangent Error loss. Results show the proposed model achieves lower Relative Percentage Difference (RPD) than baselines across low, medium, and high complexity scenarios (e.g., 36/37/38\% RPD versus several RNN baselines), demonstrating robustness to temporal irregularity and effective multi-vessel mobility modeling; the work also sets the stage for extensions with Graph Neural Networks and data fusion for global AIS streams.

Abstract

The oceans are a source of an impressive mixture of complex data that could be used to uncover relationships yet to be discovered. Such data comes from the oceans and their surface, such as Automatic Identification System (AIS) messages used for tracking vessels' trajectories. AIS messages are transmitted over radio or satellite at ideally periodic time intervals but vary irregularly over time. As such, this paper aims to model the AIS message transmission behavior through neural networks for forecasting upcoming AIS messages' content from multiple vessels, particularly in a simultaneous approach despite messages' temporal irregularities as outliers. We present a set of experiments comprising multiple algorithms for forecasting tasks with horizon sizes of varying lengths. Deep learning models (e.g., neural networks) revealed themselves to adequately preserve vessels' spatial awareness regardless of temporal irregularity. We show how convolutional layers, feed-forward networks, and recurrent neural networks can improve such tasks by working together. Experimenting with short, medium, and large-sized sequences of messages, our model achieved 36/37/38% of the Relative Percentage Difference - the lower, the better, whereas we observed 92/45/96% on the Elman's RNN, 51/52/40% on the GRU, and 129/98/61% on the LSTM. These results support our model as a driver for improving the prediction of vessel routes when analyzing multiple vessels of diverging types simultaneously under temporally noise data.

Figures (11)

• Figure 1: A cylindrical-projected map depicting the region that comprises every trajectory in the dataset of AIS messages. The region is a bounding-box from coordinates 23°52'14.8"N 82°46'58.2"W and 68°30'02.8"N 2°01'18.4"W. The trajectories in the dataset were collected between March and July 2020.
• Figure 2: A kernel-based Edge Bundling visualization technique Moura2015:3DHEB applied over the dataset's first and last message of each unique trajectory. The colors are arbitrarily used to contrast the flows and ease the visualization, while the thickness of the edges represents the flow intensity.
• Figure 3: Probability distribution of Automatic Identification System (AIS) messages per trajectory ( i.e., vessel) in the dataset. It shows that most vessels have few records, and a few vessels concentrate most of the records within the dataset, a behavior comparable to a long-tail ( i.e., Pareto) data distribution.
• Figure 4: Interquartile Range Analysis -- IQR of delta time for fifteen different vessels ordered from the one with the most AIS messages to the one with the least. The analysis reveals that all vessels present a severe presence of outliers. An outlier indicates an irregularity related to the time elapsed between two consecutive messages, varying from a couple of seconds to a few months.
• Figure 5: Temporal sampling technique designed to increase the variability of trajectories seen by the models. It decreases the computational training time on the entire trajectories without increasing the timing irregularity within the windows, proving segments of different trajectories and timespans.