The accurate and robust prediction of vessel traffic flow is gaining importance in maritime intelligent transportation system (ITS), such as vessel traffic services, maritime spatial planning, and ...traffic safety management, etc. To achieve fine-grained vessel traffic flow prediction, we will first generate the maritime traffic network (which is essentially a graph), and then propose a graph-driven neural network. In particular, to represent various correlations among spatio-temporal vessel traffic flow, we tend to extract the feature points (i.e., starting, way and ending points) by utilizing the knowledge of vessel positioning data. These feature points are essentially related to the geometrical structures of massive vessel trajectories collected from massive automatic identification system (AIS) records, contributing to the generation of maritime traffic network. We then propose a spatio-temporal multi-graph convolutional network (STMGCN)-based vessel traffic flow prediction method by exploiting multiple types of inherent correlations in the generated maritime graph. The proposed STMGCN mainly contains one spatial multi-graph convolutional layer and two temporal gated convolutional layers, beneficial for extracting spatial and temporal traffic flow patterns. The main benefit of our graph-driven prediction method is that it takes full advantage of the maritime graph and multi-graph learning. Comprehensive experiments have been implemented on realistic AIS dataset to compare our method with several state-of-the-art prediction methods. The fine-grained prediction results have demonstrated our superior performance in terms of both accuracy and robustness.
•The ship encounter identification model is formed based on AIS historical data.•A two-stage collision avoidance behavior extraction algorithm is constructed to obtain the collision avoidance ...scheme.•A novel path planning method is established by fusion the collision avoidance trajectory in similar scenarios.
AIS data include ship spatial-temporal and motion parameters which can be used to excavate the deep-seated information. In this article, an interpretable knowledge-based decision support method is established to guide the ship to make collision avoidance decisions with good seamanship and ordinary practice of seamen using AIS data. First, AIS data is preprocessed and trajectory reconstructed to restore the ship historical navigation state, and a ship encounter identification model is constructed according to the encounter characteristics; Second, a two-stage collision avoidance behavior extraction algorithm is formed to build a behavior knowledge base, and the scenario similarity model is constructed to measure and match similar scenarios based on ship position, motion tendency and collision risk. Then, the Delaunay Triangulation Network is used to fuse ship trajectories of similar scenario to form the collision avoidance path. Finally, a case study is performed using the real AIS data outside Ningbo-Zhoushan Port waters, China, and the effectiveness of the planned path is verified by setting the head-on and crossing situations and comparison between the planned and real paths. Results indicate that the proposed model can extract the ship collision avoidance behavior accurately, and the planned path can ensure navigation safety.
•Ship escort and convoy operations in ice conditions are investigated.•Ship domain size and convoy speed are in focus.•AIS data and sea ice model data is combined and processed for empirical ...analysis.•Insights in icebreaker assistance operations for different ice conditions are obtained.
Winter navigation is a complex but common operation in the Northern Baltic Sea areas. In Finnish waters, the safety of the wintertime maritime transportation system is managed through the Finnish–Swedish winter navigation system. This system results in different operational modes of ship navigation, with vessels either navigating independently or under icebreaker assistance. A recent risk analysis indicates that during icebreaker assistance, convoys operations are among the most hazardous, with convoy collisions the most important risk events. While the accident likelihood per exposure time is rather low, accidents occur almost every winter. Even though these typically lead to less serious consequences, accidents leading to ship loss and oil pollution have occurred and may occur in the future. One aspect of ship convoy navigation in ice conditions is the distance kept between the icebreaker and the ships in the convoy, a form of the well-known ship domain concept. While operational experience naturally is a valuable source of information for decision making about the distance of navigation in convoys, systematic analyses are lacking. The aim of this paper is to investigate selected operational aspects of convoy navigation in ice conditions in the Finnish waters of the Gulf of Finland, based on data of the Automatic Identification System and sea ice hindcast data. Focus is on obtaining qualitative and quantitative knowledge concerning distances between vessels in escort and convoy operations and the respective transit speeds, conditional to ice conditions. Such empirical knowledge can support operational decision making, contributing to wintertime maritime safety.
Future generation communication systems, such as 5G and 6G wireless systems, exploit the combined satellite-terrestrial communication infrastructures to extend network coverage and data throughput ...for data-driven applications. These ground-breaking techniques have promoted the rapid development of Internet of Things (IoT) in maritime industries. In maritime IoT applications, intelligent vessel traffic services can be guaranteed by collecting and analyzing high volume of spatial data flows from automatic identification system (AIS). This AIS system includes a highly integrated automatic equipment, including functionalities of core communication, tracking, and sensing. The increased utilization of shipboard AIS devices allows the collection of massive trajectory data. However, the received raw AIS data often suffers from undesirable outliers (i.e., poorly tracked timestamped points for vessel trajectories ) during signal acquisition and analog-to-digital conversion. The degraded AIS data will bring negative effects on vessel traffic services (e.g., maritime traffic monitoring, intelligent maritime navigation, vessel collision avoidance, etc.) in maritime IoT scenarios. To improve the quality of vessel trajectory records from AIS networks, we propose to develop a two-phase data-driven machine learning framework for vessel trajectory reconstruction. In particular, a density-based clustering method is introduced in the first phase to automatically recognize the undesirable outliers. The second phase proposes a bidirectional long short-term memory (BLSTM)-based supervised learning technique to restore the timestamped points degraded by random outliers in vessel trajectories. Comprehensive experiments on simulated and realistic data sets have verified the dominance of our two-phase vessel reconstruction framework compared to other competing methods. It thus has the capacity of promoting intelligent vessel traffic services in 6G-enabled maritime IoT systems.
Maritime transportation plays an essential role in global trade. Due to the huge number of vessels worldwide, there is also a non-negligible volume of Maritime incidents such as collisions/sinking ...and illegal events (e.g., piracy, smuggling, and unauthorized fishing). Electronic equipment/systems, such as radars and Automatic Identification Systems (AIS), have contributed to improving maritime situational awareness. AIS provides one of the fundamental sources of vessel kinematics and static data. Today, many approaches are focused on automatically detecting the vessels’ traffic behavior and discovering useful patterns and deviations from those data. These studies contribute to detecting suspicious activities and anomalous trajectories, whose developed techniques could be applied in the surveillance systems, helping the authorities to anticipate proper actions. Several concerns and difficulties are involved in the analyses of vessel kinematics data: how to deal with big data generated, inconsistencies, irregular updates, dynamic data, unlabeled data, and evaluation. This article presents the approaches, constraints, and challenges in maritime traffic anomaly detection research, presenting a review, a taxonomy, and a discussion of the proposed approaches.
•Several concerns are involved in the analyses of vessel kinematics data.•Manually detecting anomalous behavior is a tedious and error-prone task.•Anomaly detection is a fundamental task in different domains of knowledge.•In the maritime domain, the nonconforming patterns are referred to as anomalies.
Abstract
Understanding the distribution of fishing activity is fundamental to quantifying its impact on the seabed. Vessel monitoring system (VMS) data provides a means to understand the footprint ...(extent and intensity) of fishing activity. Automatic Identification System (AIS) data could offer a higher resolution alternative to VMS data, but differences in coverage and interpretation need to be better understood. VMS and AIS data were compared for individual scallop fishing vessels. There were substantial gaps in the AIS data coverage; AIS data only captured 26% of the time spent fishing compared to VMS data. The amount of missing data varied substantially between vessels (45–99% of each individuals' AIS data were missing). A cubic Hermite spline interpolation of VMS data provided the greatest similarity between VMS and AIS data. But the scale at which the data were analysed (size of the grid cells) had the greatest influence on estimates of fishing footprints. The present gaps in coverage of AIS may make it inappropriate for absolute estimates of fishing activity. VMS already provides a means of collecting more complete fishing position data, shielded from public view. Hence, there is an incentive to increase the VMS poll frequency to calculate more accurate fishing footprints.
The marine vessel Automatic Identification System (AIS) broadcast is a system with no-response mechanism, so there are missing data from the received AIS vessel trajectory. Currently, the mainstream ...vessel trajectory curve fitting method is achieved by means of deep learning algorithm cyclic training, however, the quality of the original data has a large impact on the trajectory fitting results, and the current method does not consider the interference of missing trajectory points on the fitted trajectory curve. Therefore, based on the data features of AIS, this paper proposes a lightweight deep learning method: Forward Backward Bidirectional Gated Recurrent Unit (FB-BiGRU). The method in this paper consists of a forward Gated Recurrent Unit (GRU) network and a backward Bidirectional Gated Recurrent Unit (BiGRU) network, reducing the range of the trajectory to be fitted by scaling in two directions simultaneously, thus gradually realizing the trajectory fitting function. The characteristics of our method are that training the trajectory data by bi-directional scaling can maximize the linear features of the trajectory curves to enhance the accuracy and efficiency of linear regression with no additional computing resources. The trajectory fitting performance of this method is verified on actual Denmark trajectory datasets, we prove that the accuracy of our proposed method in fitting short-term trajectories has increased by 49.16% and 29.89% on average compared with the (Long Short-Term Memory) LSTM and BiGRU. Furthermore, the average fitting accuracy of our method is 96 m, and the minimum fitting error is 64 m.
•A vessel trajectory fitting method based on the BiGRU network is proposed.•A BiGRU network with GRU composite model is constructed to train a trajectory fitting model.•A trajectory fitting prototype is implemented to fit vessel trajectory.•Experiments demonstrate the proposed method performs better than other methods.
A new exhaust emission inventory of ocean-going vessels (OGVs) was compiled for Hong Kong by using Automatic Identification System (AIS) data for the first time to determine typical main engine load ...factors, through vessel speed and operation mode characterization. It was found that in 2007, container vessel was the top emitting vessel type, contributing 9,886, 11,480, 1,173, 521 and 1166 tonnes of SO2, NOx, PM10, VOC and CO, respectively, or about 80%–82% of the emissions. The top five, which also included ocean cruise, oil tanker, conventional cargo vessel and dry bulk carrier, accounted for about 98% of emissions. Emission maps, which add a new spatial dimension to the inventory, show the key emission hot spots in Hong Kong and suggest that a significant portion of emissions were emitted at berth. Scientific evidence about the scale and distribution of ship emissions has contributed in raising public awareness and facilitating stakeholder engagement about the issue. Fair Winds Charter, the world's first industry-led voluntary emissions reduction initiative, is a perfect example of how careful scientific research can be used in public engagement and policy deliberation to help drive voluntary industry actions and then government proposals to control and regulate marine emissions in Hong Kong and the Pearl River Delta region.
► A detail activity-based marine vessels emission inventory for Hong Kong. ► Automatic Identification System (AIS) data used to determine main engine load factors. ► Container vessel was top emitter in 2007, contributed about 80% of marine emissions. ► Emission maps identify at-berth emission hot-spots for effective control policies. ► Fair Winds Charter: a case of industry action and policy change driven by science.
Automatic identification system (AIS) is a maritime communication system that uses a transceiver to automatically exchange navigational data. These data help navigation and allow to monitor the ...maritime traffic. However, this system can be hacked and malicious users can easily transmit false data to mislead the coastguards or vessels navigating around. While previous research has proposed methods to detect these falsifications, none of them suggest strategies that detect AIS identity spoofing combining the tracking of the ship position and AIS transceiver’s carrier frequency offset (CFO). The CFO, caused by the carrier frequencies mismatch between emitter and receiver and Doppler effect, is used as a radiometric signature to identify materially every transceiver independently of its transmitted identity. It can drift over time and this is why it is tracked thanks to a Kalman filter (KF). In addition, position is also considered to reduce the miss probability of spoofing detection. The KF is noise adaptive to be robust against various CFO drifts and noise levels of the environment. The strategy is tested on real AIS data and the results demonstrate its efficacy: false alarm and miss probabilities were respectively 1% and 1.7%. These results show the ability of the test to correctly detect identity spoofing and the interest of CFO as a radiometric signature. This signature, used for the first time in an AIS application, could be used with other signatures in a future work to improve identity spoofing detection. This is why we made open source in GitHub our algorithm and the real AIS data used.