Construction of a Real-Time Ship Trajectory Prediction Model Based on Ship Automatic Identification System Data
Abstract
:1. Introduction
2. Related Work
2.1. Ship Trajectory Prediction Based on Simulation Methods
2.2. Ship Trajectory Prediction Based on Statistical Methods
2.3. Ship Trajectory Prediction Based on Machine Learning
2.4. Ship Trajectory Prediction Based on the Hybrid Method
3. Method
3.1. AIS Data Preprocessing
3.2. Ship Trajectory Clustering
3.2.1. The CDDTW Trajectory Clustering Algorithm
3.2.2. Evaluation of the CDDTW Trajectory Clustering Algorithm
- 1.
- Evaluation indicators
- 2.
- Comparison of experimental results
3.3. Construction of the RSTPM Based on LSTM
4. Adaptability Evaluation of the RSTPM in Different River Types
4.1. Nonforked River Sections
4.1.1. Effect of Batch Size on the Prediction Model
4.1.2. Effect of the Amount of Hidden Layer Neuron Nodes in LSTM on the Prediction Model
4.1.3. Effect of Step Size on the Prediction Model
4.2. Multifork River Section
5. Experiments on the RSTPM
5.1. Analysis of Results in the Nonfork River Section
5.2. Analysis of Results in Multifork River Sections
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Lin, M. The Design and Realization of the Ship’s Trajectory Clustering Prototype System Based on AIS. Master’s Thesis, Dalian Maritime University, Dalian, China, 2016. [Google Scholar]
- Sang, L.-Z.; Yan, X.-P.; Wall, A.; Wang, J.; Mao, Z. CPA calculation method based on AIS position prediction. J. Navig. 2016, 69, 1409–1426. [Google Scholar] [CrossRef]
- Mazzarella, F.; Arguedas, V.F.; Vespe, M. Knowledge-based vessel position prediction using historical AIS data. In Proceedings of the 2015 Sensor Data Fusion: Trends, Solutions, Applications (SDF), Bonn, Germany, 6–8 October 2015; IEEE: Piscataway, NJ, USA, 2015; pp. 1–6. [Google Scholar] [CrossRef]
- Ju, C.; Wang, Z.; Long, C.; Zhang, X.; Chang, D.E. Interaction-aware kalman neural networks for trajectory prediction. In Proceedings of the 2020 IEEE Intelligent Vehicles Symposium (IV), Las Vegas, NV, USA, 19 October–13 November 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1793–1800. [Google Scholar]
- Abebe, M.; Noh, Y.; Kang, Y.-J.; Seo, C.; Kim, D.; Seo, J. Ship trajectory planning for collision avoidance using hybrid ARIMA-LSTM models. Ocean Eng. 2022, 256, 111527. [Google Scholar] [CrossRef]
- Zhang, X.; Liu, G.; Hu, C.; Ma, X. Wavelet analysis based hidden Markov model for large ship trajectory prediction. In Proceedings of the 2019 Chinese Control Conference (CCC), Guangzhou, China, 27–30 July 2019; IEEE: Piscataway, NJ, USA, 2019; pp. 2913–2918. [Google Scholar] [CrossRef]
- Dalsnes, B.R.; Hexeberg, S.; Flaten, A.L.; Eriksen, B.-O.H.; Brekke, E.F. The neighbor course distribution method with Gaussian mixture models for AIS-based vessel trajectory prediction. In Proceedings of the 2018 21st International Conference on Information Fusion (FUSION), Cambridge, UK, 10–13 July 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 580–587. [Google Scholar] [CrossRef]
- Rong, H.; Teixeira, A.; Soares, C.G. Ship trajectory uncertainty prediction based on a Gaussian Process model. Ocean Eng. 2019, 182, 499–511. [Google Scholar] [CrossRef]
- Duca, A.L.; Bacciu, C.; Marchetti, A. A K-nearest neighbor classifier for ship route prediction. In Proceedings of the OCEANS 2017-Aberdeen, Aberdeen, UK, 19–22 June 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 1–6. [Google Scholar] [CrossRef]
- Liu, J.; Shi, G.; Zhu, K. Online multiple outputs least-squares support vector regression model of ship trajectory prediction based on automatic information system data and selection mechanism. IEEE Access 2020, 8, 154727–154745. [Google Scholar] [CrossRef]
- Liu, J.; Shi, G.; Zhu, K. Vessel trajectory prediction model based on AIS sensor data and adaptive chaos differential evolution support vector regression (ACDE-SVR). Appl. Sci. 2019, 9, 2983. [Google Scholar] [CrossRef]
- Gan, S.; Liang, S.; Li, K.; Deng, J.; Cheng, T. Ship trajectory prediction for intelligent traffic management using clustering and ANN. In Proceedings of the 2016 UKACC 11th International Conference on Control (CONTROL), Belfast, UK, 31 August–2 September 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–6. [Google Scholar]
- Mao, S.; Tu, E.; Zhang, G.; Rachmawati, L.; Rajabally, E.; Huang, G.-B. An automatic identification system (AIS) database for maritime trajectory prediction and data mining. In Proceedings of ELM-2016; Springer: Cham, Germany, 2018; pp. 241–257. [Google Scholar] [CrossRef]
- Zhang, Z.; Ni, G.; Xu, Y. Trajectory prediction based on AIS and BP neural network. In Proceedings of the 2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), Chongqing, China, 11–13 December 2020; IEEE: Piscataway, NJ, USA, 2020; Volume 9, pp. 601–605. [Google Scholar] [CrossRef]
- Xu, T.; Liu, X.; Yang, X. Ship Trajectory online prediction based on BP neural network algorithm. In Proceedings of the 2011 International Conference of Information Technology, Computer Engineering and Management Sciences, Nanjing, China, 24–25 September 2011; IEEE: Piscataway, NJ, USA, 2011; Volume 1, pp. 103–106. [Google Scholar] [CrossRef]
- Zhao, L. Ship Trajectory Anomaly Detection Based on AIS Data and Recurrent Neural Network. Ph.D. Thesis, Dalian Maritime University, Dalian, China, 2019. [Google Scholar]
- Wang, H. Study on Vessel Trajectory Prediction Model Based on AIS Data. Master’s Thesis, Harbin University of Science and Technology, Harbin, China, 2021. [Google Scholar]
- Yang, B. Research and Application of the Trajectory Analysis Based on AIS. Master’s Thesis, University of Electronic Science and Technology, Sichuan, China, 2018. [Google Scholar]
- Liu, C.; Li, Y.; Jiang, R.; Du, Y.; Lu, Q.; Guo, Z. TPR-DTVN: A Routing Algorithm in Delay Tolerant Vessel Network Based on Long-Term Trajectory Prediction. Wirel. Commun. Mob. Comput. 2021, 2021, 6630265. [Google Scholar] [CrossRef]
- Wang, C.; Ren, H.; Li, H. Vessel trajectory prediction based on AIS data and bidirectional GRU. In Proceedings of the 2020 International Conference on Computer Vision, Image and Deep Learning (CVIDL), Chongqing, China, 10–12 July 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 260–264. [Google Scholar] [CrossRef]
- Wang, S.; He, Z. A prediction model of vessel trajectory based on generative adversarial network. J. Navig. 2021, 74, 1161–1171. [Google Scholar] [CrossRef]
- Sun, J. Research on Prediction of Ship Trajectory Based on CNN-LSTM. Master’s Thesis, Ningbo University, Ningbo, China, 2020. [Google Scholar]
- Chen, X.; Ling, J.; Yang, Y.; Zheng, H.; Xiong, P.; Postolache, O.; Xiong, Y. Ship trajectory reconstruction from AIS sensory data via data quality control and prediction. Math. Probl. Eng. 2020, 2020, 7191296. [Google Scholar] [CrossRef]
- Murray, B.; Perera, L.P. A dual linear autoencoder approach for vessel trajectory prediction using historical AIS data. Ocean Eng. 2020, 209, 107478. [Google Scholar] [CrossRef]
- Zhang, L.; Zhu, Y.; Su, J.; Lu, W.; Li, J.; Yao, Y. A Hybrid Prediction Model Based on KNN-LSTM for Vessel Trajectory. Mathematics 2022, 10, 4493. [Google Scholar] [CrossRef]
- Schöller, F.E.; Enevoldsen, T.T.; Becktor, J.B.; Hansen, P.N. Trajectory prediction for marine vessels using historical ais heatmaps and long short-term memory networks. IFAC-PapersOnLine 2021, 54, 83–89. [Google Scholar] [CrossRef]
- Suo, Y.; Chen, W.; Claramunt, C.; Yang, S. A ship trajectory prediction framework based on a recurrent neural network. Sensors 2020, 20, 5133. [Google Scholar] [CrossRef]
- Zhang, Z.; Ni, G.; Xu, Y. Ship trajectory prediction based on LSTM neural network. In Proceedings of the 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC), Chongqing, China, 12–14 June 2020; IEEE: Piscataway, NJ, USA, 2020; pp. 1356–1364. [Google Scholar]
- Gao, M.; Shi, G.; Li, S. Online prediction of ship behavior with automatic identification system sensor data using bidirectional long short-term memory recurrent neural network. Sensors 2018, 18, 4211. [Google Scholar] [CrossRef] [PubMed]
- Wang, L.; Chen, P.; Chen, L.; Mou, J. Ship AIS trajectory clustering: An HDBSCAN-based approach. J. Mar. Sci. Eng. 2021, 9, 566. [Google Scholar] [CrossRef]
- Widyantara, I.M.O.; Hartawan, I.P.N.; Karyawati, A.A.I.N.E.; Er, N.I.; Artana, K.B. Automatic identification system-based trajectory clustering framework to identify vessel movement pattern. IAES Int. J. Artif. Intell. 2023, 12, 1–11. [Google Scholar] [CrossRef]
- Yoo, W.; Kim, T.-W. Statistical trajectory-distance metric for nautical route clustering analysis using cross-track distance. J. Comput. Des. Eng. 2022, 9, 731–754. [Google Scholar] [CrossRef]
- Li, H.; Liu, J.; Yang, Z.; Liu, R.W.; Wu, K.; Wan, Y. Adaptively constrained dynamic time warping for time series classification and clustering. Inf. Sci. 2020, 534, 97–116. [Google Scholar] [CrossRef]
- Yang, J.; Liu, Y.; Ma, L.; Ji, C. Maritime traffic flow clustering analysis by density based trajectory clustering with noise. Ocean Eng. 2022, 249, 111001. [Google Scholar] [CrossRef]
- Tang, J.; Bi, W.; Liu, F.; Zhang, W. Exploring urban travel patterns using density-based clustering with multi-attributes from large-scaled vehicle trajectories. Phys. A Stat. Mech. Its Appl. 2020, 561, 125301. [Google Scholar] [CrossRef]
- Zhao, L.; Shi, G. A trajectory clustering method based on Douglas-Peucker compression and density for marine traffic pattern recognition. Ocean Eng. 2018, 172, 456–467. [Google Scholar] [CrossRef]
- Zeng, W.; Xu, Z.; Cai, Z.; Chu, X.; Lu, X. Aircraft trajectory clustering in terminal airspace based on deep autoencoder and gaussian mixture model. Aerospace 2021, 8, 266. [Google Scholar] [CrossRef]
- Lee, J.-S.; Lee, H.-T.; Cho, I.-S. Maritime traffic route detection framework based on statistical density analysis from AIS data using a clustering algorithm. IEEE Access 2022, 10, 23355–23366. [Google Scholar] [CrossRef]
- Ansari, M.Y.; Mainuddin; Ahmad, A.; Bhushan, G. Spatiotemporal trajectory clustering: A clustering algorithm for spatiotemporal data. Expert Syst. Appl. 2021, 178, 115048. [Google Scholar] [CrossRef]
- Anh, D.T.; Thanh, L.H. An efficient implementation of k-means clustering for time series data with DTW distance. Int. J. Bus. Intell. Data Min. 2015, 10, 213. [Google Scholar] [CrossRef]
- Chen, Y. Research on Ship Trajectory Prediction Based on Data Mining. Master’s Thesis, Harbin Engineering University, Harbin, China, 2020. [Google Scholar]
- Li, Z.W.; Wang, Z.Y.; Li, Y.F.; Liu, L. Planform geometry characteristics of typical meandering rivers in Yellow River Source. J. Sediment Res. 2012, 4, 11–17. [Google Scholar] [CrossRef]
- Wang, Q. The Prediction Model for Ship Trajectory Based on AIS Data. Master’s Thesis, Guangxi University for Nationalities, Nanning, China, 2021. [Google Scholar]
Categories | Method Categories | Advantages | Disadvantages |
---|---|---|---|
Simulation Method | Exponential smoothing model (ESM) [2] | Predictions can be made with a small amount of data | Only short-term forecasts can be made |
Curvature velocity method [3] | Simple model and good real-time | Only short-term forecasts can be made | |
Statistical Methods | Kalman filter [4] | Linear, unbiased, high accuracy | Relies on raw data and cannot predict over time |
Autoregressive moving average model (ARIMA) [5] | Simple model and wide application | Requires large amounts of data and low accuracy | |
Hidden Markov model (HMM) [6] | Good state prediction of the process | Poor robustness and complex parameter settings | |
Gaussian mixture model (GMM) [7] | High accuracy in short range prediction | Vulnerable to data complexity and low utility | |
Bayesian Networks [8] | Efficient and easy to train | Vulnerable to prior probabilities and input variables | |
Machine Learning | K-nearest neighbor (KNN) [9] | Easy to implement, no parameter estimation required | Accuracy suffers when sample size is unbalanced |
Support vector machine (SVM) [10,11] | Applicable to linear and nonlinear problems | Only for dichotomous problems | |
Artificial neural network (ANN) [12] | High accuracy and error tolerance to noise | Requires large number of initial parameters and long training time | |
Extreme learning machine (ELM) [13] | No iterations for hidden layers, fast learning | May cause overfitting problems | |
Backpropagation (BP) [14,15] | Ability to learn and generalize on your own | May fall into local extremes leading to training failure | |
Deep Learning | Long short-term memory (LSTM) [16,17,18,19] | The deficiency of long-term dependence in recurrent neural network (RNN) is effectively improved. | The internal structure is relatively complex and time-consuming to calculate |
GRU [20] | Simple model, better training speed than LSTM | Cannot completely solve the gradient disappearance problem | |
GAN [21] | Can produce clearer and more realistic samples | Not suitable for handling discrete data, such as text | |
Convolutional neural network (CNN) [22] | Feature extraction can be performed automatically | Training results easily converge to local minima | |
Deep neural network (DNN) [23] | Very good nonlinear fitting ability | Difficult to train, requires a large amount of data | |
Other | Hybrid model [24,25,26,27] | Combines the advantages of multiple models | May result in an increase in calculated costs |
Study Area | Number of Raw Data | Number of AIS Data after Step (1) Processing | Number of AIS Data after Step (4) Processing |
---|---|---|---|
The Wuhan section of the dendritic river system in the Yangtze River’s middle reaches | 8,538,423 | 1,767,305 | 4,428,680 |
The partial reticulated river system in the northern part of the Zhejiang Province | 1,850,761 | 217,770 | 1,302,994 |
Algorithm | Number of Sampling Points | ||||
---|---|---|---|---|---|
100 | 500 | 1000 | 3000 | 5000 | |
DBSCAN | 0.205 | 0.940 | 3.352 | 28.84 | 81.8 |
DBSCAN + Ball-Tree | 0.142 | 0.201 | 0.298 | 0.801 | 1.533 |
Algorithm | Intra-Cluster Distance | ||||||
---|---|---|---|---|---|---|---|
Cluster 1 | Cluster 2 | Cluster 3 | Cluster 4 | Cluster 5 | Cluster 6 | Cluster 7 | |
DTW | 437 | 798 | 632 | 750 | 339 | 568 | 466 |
CDDTW | 415 | 469 | 458 | 322 | 343 | 542 | 430 |
HDBSCAN | 462 | 543 | 463 | 512 | 351 | 497 | 393 |
OPTICS | 431 | 586 | 538 | 487 | 387 | 572 | 449 |
Model Parameters | Study Area | |||||
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | |
River bending coefficient (m/m) | 1.017 | 1.269 | 1.287 | 1.343 | 1.746 | 2.493 |
Number of hidden layer neurons (pcs) | 88 | 108 | 128 | 108 | 168 | 188 |
Training data step (step) | 14 | 12 | 12 | 12 | 10 | 8 |
Predicted average accuracy (m) | 35 | 20 | 40 | 20 | 57 | 24 |
Model Parameters | Cluster | ||||||||
---|---|---|---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | |
River bending coefficient (m/m) | 1.078 | 4.509 | 1.368 | 1.342 | 1.930 | 1.156 | 1.384 | 1.152 | 1.038 |
Number of hidden layer neurons (pcs) | 128 | 188 | 108 | 108 | 168 | 168 | 108 | 108 | 128 |
Training data step (step) | 14 | 8 | 12 | 12 | 10 | 10 | 12 | 12 | 14 |
Predicted average accuracy (m) | 14 | 52 | 16 | 15 | 30 | 14 | 21 | 24 | 16 |
MMSI | Models | Distance Error (km) | Speed Error (km) | Heading Error (Degree) | |||
---|---|---|---|---|---|---|---|
Max | Mean | Max | Mean | Max | Mean | ||
413827414 | GRU | 0.0922 | 0.0269 | 0.331 | 0.069 | 6.210 | 1.130 |
RSTPM | 0.0884 | 0.0204 | 0.360 | 0.063 | 5.664 | 1.022 | |
413812757 | GRU | 0.0734 | 0.0285 | 0.526 | 0.082 | 11.524 | 1.261 |
RSTPM | 0.0758 | 0.0208 | 0.537 | 0.073 | 12.157 | 1.121 | |
413812608 | GRU | 0.0692 | 0.0266 | 0.997 | 0.078 | 7.214 | 1.267 |
RSTPM | 0.0758 | 0.0211 | 0.793 | 0.072 | 7.234 | 1.188 | |
413819268 | GRU | 0.0877 | 0.0266 | 0.531 | 0.069 | 9.041 | 1.126 |
RSTPM | 0.0581 | 0.0184 | 0.551 | 0.066 | 7.010 | 0.997 |
MMSI | Models | Distance Error (km) | Speed Error (km) | Heading Error (Degree) | |||
---|---|---|---|---|---|---|---|
Max | Mean | Max | Mean | Max | Mean | ||
413823145 | GRU | 0.0442 | 0.0171 | 0.362 | 0.098 | 5.194 | 1.433 |
RSTPM | 0.0362 | 0.0144 | 0.322 | 0.096 | 4.417 | 1.158 | |
413791454 | GRU | 0.0354 | 0.0149 | 0.293 | 0.068 | 10.852 | 2.135 |
RSTPM | 0.0354 | 0.0131 | 0.286 | 0.069 | 7.434 | 1.847 | |
413981121 | GRU | 0.0382 | 0.0154 | 0.507 | 0.110 | 5.003 | 1.577 |
RSTPM | 0.0397 | 0.0147 | 0.555 | 0.111 | 7.360 | 1.570 | |
413977486 | GRU | 0.0300 | 0.0163 | 0.213 | 0.063 | 6.292 | 2.110 |
RSTPM | 0.0285 | 0.0138 | 0.210 | 0.064 | 6.794 | 1.627 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Xi, D.; Feng, Y.; Jiang, W.; Yang, N.; Hu, X.; Wang, C. Construction of a Real-Time Ship Trajectory Prediction Model Based on Ship Automatic Identification System Data. ISPRS Int. J. Geo-Inf. 2023, 12, 502. https://doi.org/10.3390/ijgi12120502
Xi D, Feng Y, Jiang W, Yang N, Hu X, Wang C. Construction of a Real-Time Ship Trajectory Prediction Model Based on Ship Automatic Identification System Data. ISPRS International Journal of Geo-Information. 2023; 12(12):502. https://doi.org/10.3390/ijgi12120502
Chicago/Turabian StyleXi, Daping, Yuhao Feng, Wenping Jiang, Nai Yang, Xini Hu, and Chuyuan Wang. 2023. "Construction of a Real-Time Ship Trajectory Prediction Model Based on Ship Automatic Identification System Data" ISPRS International Journal of Geo-Information 12, no. 12: 502. https://doi.org/10.3390/ijgi12120502