Adaptive reconstruction of ship trajectory considering motion states

A method for simplifying ship trajectory based on improved Douglas–Peucker algorithm

A novel ship trajectory reconstruction approach based on low-rank tensor completion

Accurate ship trajectory plays an important role for maritime traffic control and management, and ship trajectory prediction with Automatic Identification System (AIS) data has attracted considerable research attentions in maritime traffic community. The raw AIS data may be contaminated by noises, which limits its usage in maritime traffic management applications in real world. To address the issue, we proposed an ensemble ship trajectory reconstruction framework combining data quality control procedure and prediction module. More specifically, the proposed framework implemented the data quality control procedure in three steps: trajectory separation, data denoising, and normalization. In greater detail, the data quality control procedure firstly identified outliers from the raw ship AIS data sample, which were further cleansed with the moving average model. Then, the denoised data were normalized into evenly distributed data series (in terms of time interval). After that, the proposed framework predicted ship trajectory with the artificial neural network. We verified the proposed model performance with two ship trajectories downloaded from public accessible AIS data base.

Background/Objectives: Ship trajectories in Vessel Traffic Service (VTS) system are generated by integrating the Automatic Identification System (AIS) or Radar system. However, the AIS system has missing data section caused by AIS device problems, radio jamming, and so on. These data have been confusing ship navigators and VTS operators.Methods/Statistical analysis: In order to extract missing AIS data, time intervals of sequent points from each ship trajectory are calculated. The section with missing AIS data is above a threshold time limit defined by characteristics. Using k-means algorithm, missing AIS data were clustered into several clusters stored by ship’s ID and sailing direction. Using association rule mining analysis, meaningful association pattern were calculated by missing AIS dataset.Findings: As a result of the association rule mining, we found several missing AIS situation patterns. In case of the west route, the probability of missing AIS situation is high when they enter the east and passenger routes. Also, the probability of missing AIS situation of passing the passenger route is high when that ship enter the LNG, east and west routes.Improvements/Applications: These results can be used to predict the probability of missing AIS data in VTS system.

A maritime traffic route extraction method based on density-based spatial clustering of applications with noise for multi-dimensional data

Automatic identification system (AIS) data record a ship’s position, speed over ground (SOG), course over ground (COG), and other behavioral attributes at specific time intervals during a ship’s voyage. At present, there are few studies in the literature on ship trajectory classification, especially the clustering of trajectory segments, to measure the multi-dimensional information of trajectories. Therefore, it is necessary to fully utilize the multi-dimensional information from AIS data when utilizing ship trajectory classification methods. Here, we propose a ship trajectory classification method based on multi-attribute trajectory similarity metrics which utilizes the following steps: (1) Improve the Douglas–Peucker (DP) algorithm by considering the SOG and COG; (2) use a multi-attribute symmetric segmentation path distance (MSSPD) for the similarity metric between trajectories; (3) cluster the segmented sub-trajectories based on the density-based spatial clustering of applications with noise (DBSCAN) algorithm; (4) adaptively determinate the optimal input parameters based on the proposed comprehensive clustering performance metrics. The proposed method was tested on real AIS data from Bohai Sea waters, and the experimental results show that the algorithm can accurately cluster the ship trajectory groups and extract traffic distributions in key waters.

High-quality ship trajectory clustering is an important fundament for learning meaningful knowledge from massive historical trajectories, which are commonly extracted from the shipborne Automatic Identification System (AIS) data. However, clustering of massive trajectory data easily suffers from two limitations, i.e., low accuracy and high computational cost. In this work, we propose to develop an efficient ship trajectory clustering method based on the virtual maritime traffic network, which is constructed using the massive AIS data. In particular, we first adopt the popular Douglas-Peucker (DP) algorithm to simplify the ship trajectories to obtain the first type of feature points, where the moving ships change the course significantly. The second type of features points, where exist the large number of ship encounters, is also obtained through estimating the ship density map. The density-based clustering method is employed to remove the unwanted outliers in these feature points, contributing to the extraction of important waypoints. The virtual maritime traffic network is then constructed using the extracted way-points. To obviously shorten the computational cost, the virtual network is finally used to cluster the massive ship trajectories by introducing the concepts of base cluster and stream cluster. Two famous metrics, i.e., Silhouette Coefficient (SC) and Davies-Bouldin Index (DBI), will be exploited to evaluate the trajectory clustering results for our method and several competing methods. Experimental results on realistic AIS data have demonstrated the superior clustering performance of our method in terms of computational efficiency, SC and DBI metrics.

TTMRN: A topological-geometric two-layer maritime route network modeling for ship intelligent navigation

In recent years, the prediction of ship trajectory based on automatic identification system (AIS) data has become an important area of research. Among the existing studies, most focus on a single ship to extract features and train models for trajectory prediction. However, in a real situation, AIS contains a variety of ships and trajectories that need a general model to serve various cases. Therefore, in this paper, we include an attentional mechanism to train a multi-trajectory prediction model. There are three major processes in our model. Firstly, we improve the traditional density-based spatial clustering of applications with noise (DBSCAN) algorithm and apply it to trajectory clustering. According to the clustering process, ship trajectories can be automatically separated by groups. Secondly, we propose a feature extraction method based on a hierarchical clustering method for a trajectory group. According to the extraction process, typical trajectories can be obtained for individual groups. Thirdly, we propose a multi-trajectory prediction model based on an attentional mechanism. The proposed model was trained using typical trajectories and tested using original trajectories. In the experiments, we chose nearby port waters as the target, which contain various ships and trajectories, to validate our model. The experimental results show that the mean absolute errors (MAEs) of the model in longitude (°) and latitude (°) compared with the baseline methods were reduced by 8.69% and 6.12%.

RAGAN: A Generative Adversarial Network for risk-aware trajectory prediction in multi-ship encounter situations

Vessel trajectory information plays a critical role in vessel traffic management systems by enhancing navigational safety. However, trajectory reconstruction remains challenging due to the inherent noise and incompleteness in automatic identification system (AIS) measurements. In this paper, we propose non-homogeneous Markov chain driven interacting multiple model (NHMC-IMM), which is a novel adaptive information fusion approach by integrating multiple motion models to adapt to the diversity of target motion. The gradient descent algorithm is applied to the improved IMM, which is used to adjust the covariance matrix of the process noise and the measurement noise during the execution of the IMM. In this way, the accuracy of the model estimation is improved by updating the parameters continuously and iteratively under the given conditions of the observed measurement data, and the optimal parameters will be selected at the end of the iteration, thus enhancing the accuracy of trajectory reconstruction. To validate the effectiveness of the proposed method, experiments were conducted using AIS datasets from the coastal regions of the United States and Denmark, under varying levels of local measurement missingness. Results demonstrate superior accuracy and robustness in trajectory reconstruction, highlighting the method’s potential for improving maritime measurement systems and enhancing situational awareness.

Ship trajectory prediction is crucial for collision avoidance and maritime traffic management. In an encounter scenario, a ship must predict the future states of other ships to take effective collision avoidance actions. However, conventional models often fail to accurately capture the interactions between ships in converging waters, resulting in poor trajectory prediction. This paper proposes a new model, STETC, which combines a Transformer and a convolutional neural network (CNN) into a two-layer encoder–decoder structure. The motion path encoder employs a self-attention mechanism to extract motion features from historical trajectories. The encounter interaction encoder integrates a CNN and self-attention mechanisms to extract interaction features from an artificial potential field generated by ship dynamic parameters. The two decoders then use cross-attention to progressively establish the spatiotemporal relationships between motion and interaction features. Through training, the model learns the interaction patterns between ships and the dynamic development of encounter situations. Real ship trajectory data is used to validate the effectiveness of STETC. Comparison experiments with observation times of 5, 10, and 15 min demonstrate that the STETC model outperformed other models. The case study validates that the STETC model can precisely perceive the motion states of surrounding ships and, leveraging encounter scenarios, generates predicted trajectories that reflect the interactive behaviors of the ships more accurately.

Ship collision avoidance behaviour recognition and analysis based on AIS data

선박교통류 평가는 대상해역 해상교통특성 파악을 통하여 정량화된 체계화된 교통관리를 수행함으로써 해양사고를 예방하는데 중요하다. 해상교통류의 특성은 선박자동식별장치(AIS, Automatic Identification System)가 선박에 설치됨에 따라 이 장치에서 수신된 정보를 토대로 평가할 수 있다. 본 연구에서는 해상교통관제센터(VTS, Vessel Traffic Services)에서 선박으로부터 수집된 위치, 속력, 침로, 시간 정보로부터 항로상에 해상교통류를 평가하기 위한 시뮬레이터를 구현하였다. 그 결과 해상교통관제센터에서 수집된 데이터로부터 선박 통항량, 통항분포, 선박속력분포, 기하학적 충돌빈도가 계산되어 표시된다. 이러한 해상교통류평가 평가는 해상교통관제사가 해상교통관리의 관점에서 해상안전을 위한 서비스를 제공하는데 유효하게 활용될 수 있을 것으로 사료된다. Maritime traffic assessment is important to understand the characteristics of maritime traffic and to prevent maritime accidents. The maritime traffic assessment can be calculated from the ship trajectory data observed by using AIS(Automatic Identification System). This paper developes a maritime traffic assessment tool using ship's position and speed, course, time data from ships navigating waterways. The results are represented in terms of the number of traffic quantity and traffic distribution, speed distribution, geometric collision candidates. The developed tool will contributes to advance maritime traffic safety by VTS(Vessel Traffic Services).

With the enormous amount of information provided by the ship Automatic Identification System (AIS), AIS is now playing a significant role in maritime transport system-related research and development. Many kinds of research and industrial applications are based on the ship trajectory extracted from raw AIS data. However, due to the issues of equipment, the transmission environment, and human factors, the raw AIS data inevitably contain abnormal messages, which have hindered the utilization of such information in practice. Thus, in this paper, an anomaly detection method that focuses on AIS trajectory is proposed, making comprehensive use of the kinematic information of the ship in the AIS data. The method employs three steps to obtain non-error AIS trajectories: (1) data preprocessing, (2) kinematic estimation, and (3) error clustering. It should be noted that steps (2) and (3) are involved in an iterative process to determine all of the abnormal data. A case study is then conducted to test the proposed method on real-world AIS data, followed by a comparison between the proposed method and the rule-based anomaly detection method. As the processed trajectories show fewer abnormal features, the results indicate that the method improves performance and can accurately detect as much abnormal data as possible.