The detection of unsafe ship behaviors in bridge waterways is crucial for ship collision avoidance. To enhance navigation safety, this study proposes a novel model that integrates a heuristic algorithm with a neural network for behavior detection. First, this study conducts an in-depth analysis of AIS data from consecutive bridge zones in the Wuhan section of the Yangtze River. In accordance with maritime regulations, four types of unsafe ship behaviors are defined: overspeeding, turning, crossing, and anchoring. Then, through an examination of multi-dimensional features, such as ship position, speed, and heading, a thorough analysis of navigation patterns and rules is conducted. This analysis yields specific decision criteria for the unsafe behaviors, thus establishing a specialized database for ship behaviors in bridge waterways. Finally, an improved PSO-LSTM model is developed to mitigate the negative impact of random parameter initialization in conventional LSTM networks on detection accuracy. Experimental results indicate that the proposed model exhibits clear superiority over SVM, BP, LSTM, and improved PSO-BP models in both visualization and evaluation metrics, achieving high detection precision for unsafe ship behaviors in bridge waterways and thus providing reliable decision-making support for maritime authorities.
HANDAYANID O D, SEDIONOW, SHAHA .Identification of vessel anomaly behavior using support vector machines and Bayesian networks[C]//2014 International Conference on Computer and Communication Engineering,September 23-25,2014,Kuala Lumpur,Malaysia. IEEE,2015: 258-261.
[2]
PALLOTTAG, VESPEM, BRYANK .Vessel pattern knowledge discovery from AIS data:a framework for anomaly detection and route prediction[J].Entropy,2013,15(6):2218-2245.
[3]
RIVEIROM J. Visual analytics for maritime anomaly detection[J]. Visual Analytics for Maritime Anomaly Detection, 2011, 46:208.
[4]
PIPANMEKAPORNL, KAMONSANTIROJS .A deep learning approach for fishing vessel classification from VMS trajectories using recurrent neural networks[C]//Human Interaction,Emerging Technologies and Future Applications Ⅱ.Cham:Springer,2020:135-141.
YANGF, HEZ W, HEF. Detection method of abnormal ship behavior based on LSTM neural network[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2019, 43(5): 886-892.(in Chinese)
WANGL L, LIUJ. Ship behavior recognition method based on multi-scale convolution[J]. Journal of Computer Applications,2019, 39(12): 3691-3696.(in Chinese)
[9]
宁耀 .基于深度学习的渔船行为识别方法研究[D].兰州:兰州大学,2020.
[10]
NINGY. Research on the behavior identification method of fishing vessels based on deep learning[D]. Lanzhou University,2020.(in Chinese)
[11]
FENGY, ZHAOX L, HANM X,et al .The study of identification of fishing vessel behavior based on VMS data[C]//Proceedings of the 3rd International Conference on Telecommunications and Communication Engineering,November 9-12,2019,Tokyo,Japan. ACM, 2020: 63-68.
NGUYEND, SIMONINM, HAJDUCHG,et al .Detection of abnormal vessel behaviours from AIS data using GeoTrackNet:from the laboratory to the ocean[C]//2020 21st IEEE International Conference on Mobile Data Management (MDM),June 30 - July 3,2020,Versailles,France.IEEE,2020:264-268.
[15]
KENNEDYJ, EBERHARTR .Particle swarm optimization[C]//Proceedings of ICNN'95-International Conference on Neural Networks, November 27-December 1,1995,Perth,WA,Australia, IEEE,2002:1942-1948.
[16]
TANGZ Y, ZHANGD X .A modified particle swarm optimization with an adaptive acceleration coefficients[C]//2009 Asia-Pacific Conference on Information Processing,July 18-19,2009,Shenzhen,China. IEEE,2009:330-332.
[17]
SHIY, EBERHARTR. A modified particle swarm optimizer[C] //1998 IEEE International Conference on Evolutionary Computation Proceedings May 4-9, 1998, Anchorage, AK, USA. IEEE, 1998. 69-73.
[18]
QIANL, ZHENGY Z, LIL,et al .A new method of inland water ship trajectory prediction based on long short-term memory network optimized by genetic algorithm[J].Applied Sciences,2022,12(8):4073.
ZHENGY Z, LIX, QIANL,et al .Ship violation behavior detection based on DCNN-LSTM model[J].Journal of Hunan University (Natural Sciences),2024,51(12):119-128.(in Chinese)
SUNL. Introduction to the self-turned berthing methods and key points of small ships[C]//2015 China Maritime Day Forum and 2015 China Pilotage Development Forum. 2015:216-218. (in Chinese).
LIUZ Y, ZHOUC H, SUNY F,et al .Identification and excavation method of ship anchoring behavior in Minjiang Estuary waters[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering),2021,45(4):805-810.(in Chinese)
CUIH H .Development of towing cable tension monitoring and auxiliary decision system based on.NET[D].Dalian:Dalian Maritime University,2020.(in Chinese)