Reliable anti-collision control algorithms conforming with the rules regulating traffic at sea, the International Regulations for Preventing Collisions at Sea (COLREG), are essential for the deployment of autonomous vessels in waters shared with other ships. The development of such methods is an active field of research. However, little attention has been given to how these rules are interpreted by experienced mariners, and how such information can be parametrised for use in automatic control systems and autonomous ships. This paper presents a method for exploiting historical automatic identification system (AIS) data to characterise parameters indicating the prevalent practices at sea in encounters with high collision risk. The method has been tested on data gathered in areas off the Norwegian coast over several years. Statistics on relevant parameters from the resulting dataset and the relation between them is presented. The results indicate that the strongest influence on vessel behaviour is the type of situation, and the amount of land and grounding hazards in the vessel's proximity.

This paper introduces a novel approach for extracting vessel navigation patterns from very large automatic identification system (AIS) trajectory big data. AIS trajectory data records are first converted to a series of code documents using vector quantisation, such as k-means and PQk-means algorithms, whose performance is evaluated in terms of precision and computational time. Therefore, a topic model is applied to these code documents from which vessels’ navigation patterns are extracted and identified. The potential of the proposed approach is illustrated by several experiments conducted with a practical AIS dataset in a region of North West France. These experimental results show that the proposed approach is highly appropriate for mining AIS trajectory big data and outperforms common DBSCAN algorithms and Gaussian mixture models.

Ship anomaly detection is a vital aspect for monitoring navigational safety in specific water areas. Considering the effect of water channel boundaries, we propose the detection of an abnormal ship trajectory based on the complex polygon (DATCP) method to detect ship anomalies in this study. With the automatic identification systems (AIS) data from the Yangtze River estuary, a case study is created to verify the effectiveness of the proposed DATCP method. The case study results reveal that the proposed DATCP method can provide higher detection accuracy than the conventional A* algorithm. The feature analysis results indicate that ship anomalies are significantly influenced by ship type, time period, weather conditions and ship traffic characteristics.

Developing risk models to predict where collisions between vessels might occur is hindered by the relative sparsity of collisions. To address this, vessel encounters and near-misses have been used as a surrogate indicator of collision risk, referred to as ‘domain analysis’. When constructed empirically, using historical information, previous work is challenged by the multitude of factors which influence the passing distances between vessels. Within this paper, we conduct data mining of big vessel traffic datasets to determine the encounter characteristics across different waterways, vessel types and speeds, weather conditions and other exploratory variables. To achieve this, we utilise a novel approach of machine learning through a random forest algorithm to predict the critical passing distance between vessels in a multitude of conditions. We contribute a far greater range of influencing factors on domain size and shape than previous studies. Finally, we investigate the potential advantages of this approach to assess the spatial risk of collision across a large region. The results help to establish the factors that influence collision risk between navigating vessels and enable empirical maritime risk assessments.

For maritime safety and security, vessels should be able to predict the trajectories of nearby vessels to avoid collision. This research proposes three novel models based on similarity search of trajectories that predict vessels' trajectories in the short and long term. The first and second prediction models are, respectively, point-based and trajectory-based models that consider constant distances between target and sample trajectories. The third prediction model is a trajectory-based model that exploits a long short-term memory approach to measure the dynamic distance between target and sample trajectories. To evaluate the performance of the proposed models, they are applied to a real automatic identification system (AIS) vessel dataset in the Strait of Georgia, USA. The models' accuracies in terms of Haversine distance between the predicted and actual positions show relative prediction error reductions of 40·85% for the second model compared with the first model and 23% for the third model compared with the second model.

Scouring and sedimentation effects on the seabed induced by ship propellers during ship manoeuvring near harbour structures affect both structure stability and ship manoeuvring capabilities. This contribution proposes solutions at an operational level using the automatic identification system (AIS) and a bridge simulator. Two new alternative manoeuvres were designed and tested on a bridge simulator to obtain expected maximum scour depth and the results were compared with that of real manoeuvres (i) using mooring lines, and (ii) with tug assistance. A total of 42 test scenarios combining several manoeuvres and meteorological conditions were reproduced. Results confirmed a clear reduction in erosion depth with the alternative manoeuvres, with total reduction when using the tugboat. The presented methodology can be very useful to port authorities to prevent the effects of ship erosion on harbour infrastructures.

In order to determine fairway width accurately and to ensure an adequate level of safety, it is necessary to fulfil different safety standards and user needs. There are a number of international recommendations for fairway design, but the recommendations by Permanent International Association of Navigation Congresses (PIANC), Puerto Del Estado (ROM 3·1) and the Japanese Ministry of Land, Infrastructure, Transport and Tourism (MLIT) are some of the best known. However, in all of these recommendations, additional width due to ship-bank effect is obsolete and needs to be improved. The objective of this paper is to use a navigational simulator with a predefined ship-bank interaction model and fine form ships models to determine the ship trajectory caused by ship-bank effect. The methodology of research used in the paper consisted of using ship trajectory data to propose a deterministic model for improving fairway additional width due to ship-bank effect. Results showed that such a method is not time-consuming and can improve fairway design in terms of ship-bank effect, increasing additional width accuracy.

When ships approach each other, they should keep a minimum area around them clear of other vessels in order to remain safe. The geometrical shape of this area has been studied since the early 1970s and is defined as the ship domain. The progress in computer capacity since then and the introduction of the Automatic Identification System (AIS) provides the potential to further investigate the size and the governing factors of the domain. This investigation revisits and proposes a method using data based on 600,000 ship encounters at 36 locations. It is concluded that the ship domain has the shape of an ellipse with half axis radii of 0.9 and 0.45 nautical miles. However, there are two factors that greatly affect the ship domain: how large the area is that is used to gather vessel intersections and whether they are constrained by water depth. In contradiction to some previous research, it is found that the ship domain is unrelated to the length of the ship.

Two pre-processes for Automatic Identification System (AIS) trajectories commonly reported in the maritime knowledge discovery literature are trajectory simplification and event identification. Both pre-processes reduce storage and computational expenses by reducing the number of data points to be used in an analysis. This paper presents an event identification and trajectory simplification method based on behaviour identification and translation. Trajectory segments deemed to correspond to coastal or ocean navigation are translated into equivalent passage plan segments; a succinct description of the movements and behaviour of the ship. As a trajectory simplification method, it provides two main advantages over commonly used trajectory simplification methods: more meaningful simplified trajectories with better encoding of basic behaviours and the possibility to retain interesting behaviours in full resolution. As an event identification method, it is capable of differentiating between normal ocean or coastal navigating behaviour and complex or interesting behaviour, such as pilotage, reaction to a traffic conflict, or an involuntary deviation from the passage plan.

Developing adequate ship domain models may significantly benefit vessel navigation safety. In essence, navigation safety is collectively affected by the navigable waterway condition, the size and shape of the ship, and operators' skills. The existing ship domains mainly use constant values for the model input parameters, making them incapable of handling site-specific conditions. This study proposes dynamic ship domain models that take into consideration navigable waterway conditions, ship behaviours, ship types and sizes, and operators' skills in a holistic manner. Specifically, the conditions of restricted waterways are classified into navigating along the channel, crossing the channel, joining another flow and turning. The ship types considered include ships that transport non-hazardous goods and Liquid Natural Gas (LNG) ships that are in need of additional security zones. A computational experiment is conducted for model application using data on water channel design and ship traffic volumes related to navigating along the channel, joining another flow and turning. Comparisons of results obtained between the proposed dynamic models with real ship traffic counts reveal that the proposed models could achieve a higher level of accuracy in estimating the capacity of restricted water channels. It therefore could potentially deliver safety enhancements of waterway transportation.