- 无标题文档
查看论文信息

中文题名:

 

基于深度学习的船舶航行轨迹预测方法研究

    

姓名:

 李岩    

学号:

 1049721801058    

保密级别:

 公开    

论文语种:

 chi    

学科代码:

 082302    

学科名称:

 工学 - 交通运输工程 - 交通信息工程及控制    

学生类型:

 硕士    

学位:

 工学硕士    

学校:

 武汉理工大学    

院系:

 航运学院    

专业:

 交通信息工程及控制    

研究方向:

 交通环境与安全保障    

第一导师姓名:

 刘文    

第一导师院系:

 武汉理工大学    

完成日期:

 2021-04-12    

答辩日期:

 2021-05-29    

中文关键词:

 

船舶航行轨迹预测 ; 深度学习 ; LSTM ; 斥力模型 ; 四元船舶领域

    

中文摘要:

随着船舶自动识别系统(Automatic Identification Systems, AIS)的强制配备和广泛使用,海量AIS数据的获取促进了船舶轨迹数据挖掘和水上交通态势感知等领域的快速发展。针对通航环境的复杂性、航行船舶的独特性和航行态势的不确定性,准确预测船舶航行轨迹已成为海事智能监管和船舶自主航行的热点和难点问题。随着深度学习技术的迅猛发展,在分析船舶行为特征和AIS大数据的基础上,构建基于深度循环神经网络的船舶航行轨迹预测模型,对于海上交通安全和智能化发展具有重要的研究意义。本文在深入分析基于深度学习的轨迹预测理论的基础上,通过引入社会力模型和船舶领域等概念,构建了基于长短期记忆网络(Long Short-Term Memory, LSTM)的船舶航行轨迹预测模型,并利用不同水域的真实AIS数据验证了其有效性和稳定性。本文主要研究内容及创新性成果包括:

(1)针对传统LSTM难以高精度预测船舶航行轨迹的问题,本文提出了基于无冲突LSTM的轨迹预测模型。该模型将每艘船舶视为一个LSTM单元,通过融合其它船舶的位置信息,实现两船之间位置信息的交互。同时,引入社会力模型中的斥力模型,计算两船之间的虚拟斥力,对避让信息进行建模,实现两船之间避让信息的交互。最后在损失函数中引入误差偏移角来提高轨迹预测的精度和稳定性。

(2)针对无冲突LSTM模型在计算两船间最小安全间距时未考虑船舶航速的问题,本文将提出融合四元船舶领域(Quaternion Ship Domain, QSD)和LSTM的船舶航行轨迹预测模型,即QSD-LSTM。该模型通过引入QSD概念来确定本船在不同运动状态下的船舶领域,同时考虑船舶航行中的航速信息,计算他船与本船船舶领域所构建的不规则椭圆相切时的临界值,进一步优化两船间最小安全距离的计算结果,以提升不同通航环境下的船舶航行轨迹预测精度。

(3)为验证两个模型的有效性,本文将在采集曹妃甸、成山角及舟山群岛水域真实AIS数据的基础上,利用平均偏移误差、终点偏移误差及平均非线性位置偏移误差等指标来综合评价船舶航行轨迹预测结果。实验结果表明,本文提出的无冲突LSTM相较于传统LSTM能够实现两船之间位置及避让信息的交互,可获取更高精度的船舶航行轨迹预测结果;为实现两船之间交互过程的动态建模,进一步在无冲突LSTM中引入QSD概念,得到的QSD-LSTM相比无冲突LSTM和传统LSTM有着稳定性更强、精度更高的预测结果。

针对原始LSTM对船舶航行轨迹预测能力不足的问题,本文构建了无冲突LSTM预测模型,该模型实现了两船之间位置及避让信息的交互。为进一步优化无冲突LSTM模型,提出了QSD-LSTM预测模型,该模型在训练中实现了两船之间避让信息的动态交互。本文的相关研究成果能够实现不同通航环境下船舶航行轨迹的高精度预测,为海事智能监管和船舶自主航行等领域提供了技术支撑。

参考文献:

[1]刘长俭. 交通强国背景下中国水运发展战略思考[J]. 中国水运, 2019(12):14-15.

[2]2020-2026年中国无人水面艇行业调研与市场前景报告[R]. 中国市场调研网, 2019.

[3]高雅. 移动对象轨迹数据的位置预测[D]. 南京: 南京航空航天大学, 2019.

[4]Vashishtha D, Panda M. Maximum likelihood multiple model filtering for path prediction in intelligent transportation systems[J]. Procedia computer science, 2018, 143: 635-644.

[5]乔少杰, 韩楠, 朱新文, 等. 基于卡尔曼滤波的动态轨迹预测算法[J]. 电子学报, 2018, 46(2): 418-423.

[6]Cartlidge J, Gong S, Bai R, et al. Spatio-temporal prediction of shopping behaviours using taxi trajectory data[C]//2018 IEEE 3rd International Conference on Big Data Analysis (ICBDA). IEEE, 2018: 112-116.

[7]高建, 毛莺池, 李志涛. 基于高斯混合一时间序列模型的轨迹预测[J]. 计算机应用, 2019, 39(8): 2261-2270.

[8]Qiao S, Shen D, Wang X, et al. A self-adaptive parameter selection trajectory prediction approach via hidden Markov models[J]. IEEE Transactions on Intelligent Transportation Systems, 2014, 16(1): 284-296.

[9]Asahara A, Maruyama K, Sato A, et al. Pedestrian-movement prediction based on mixed Markov-chain model[C]//Proceedings of the 19th ACM SIGSPATIAL international conference on advances in geographic information systems. 2011: 25-33.

[10]Dalsnes B R, Hexeberg S, Flaten A L, et al. The neighbor course distribution method with Gaussian mixture models for AIS-based vessel trajectory prediction[C]//2018 21st International Conference on Information Fusion (FUSION). IEEE, 2018: 580-587.

[11]乔少杰, 金琨, 韩楠, 等. 一种基于高斯混合模型的轨迹预测算法[J]. 软件学报, 2015, 26(5): 1048-1063.

[12]Wiest J, Hoffken M, Krepel U, et al. Probabilistic trajectory prediction with gaussian mixture models[C]//2012 IEEE Intelligent Vehicles Symposium. IEEE, 2012: 141-146.

[13]Lim Q, Johari K, Tan U X. Gaussian Process Auto Regression for vehicle center coordinates Trajectory Prediction[C]//TENCON 2019-2019 IEEE Region 10 Conference (TENCON). IEEE, 2019: 25-30.

[14]王新立. 基于GPR模型的船舶自适应轨迹预测及应用研究[D]. 武汉: 武汉理工大学, 2019.

[15]乔少杰, 彭京, 李天瑞, 等. 基于CTBN的移动对象不确定轨迹预测算法[J]. 电子科技大学学报, 2012, 41(5): 759-763.

[16]He G, Li X, Lv Y, et al. Probabilistic intention prediction and trajectory generation based on dynamic bayesian networks[C]//2019 Chinese Automation Congress (CAC). IEEE, 2019: 2646-2651.

[17]McCulloch W S, Pitts W. A logical calculus of the ideas immanent in nervous activity[J]. The bulletin of mathematical biophysics, 1943, 5(4): 115-133.

[18]Gaxiola F, Melin P, Valdez F, et al. Interval type-2 fuzzy weight adjustment for backpropagation neural networks with application in time series prediction[J]. Information Sciences, 2014, 260: 1-14.

[19]Chen Y, Yang B, Dong J. Time-series prediction using a local linear wavelet neural network[J]. Neurocomputing, 2006, 69(4-6): 449-465.

[20]Tang J, Liu F, Zou Y, et al. An improved fuzzy neural network for traffic speed prediction considering periodic characteristic[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(9): 2340-2350.

[21]Chandra R. Competition and collaboration in cooperative coevolution of Elman recurrent neural networks for time-series prediction[J]. IEEE transactions on neural networks and learning systems, 2015, 26(12): 3123-3136.

[22]Specht D F. A general regression neural network[J]. IEEE transactions on neural networks, 1991, 2(6): 568-576.

[23]Han H G, Chen Q, Qiao J F. An efficient self-organizing RBF neural network for water quality prediction[J]. Neural Networks, 2011, 24(7): 717-725.

[24]Abu-Mostafa Y, Jacques J S. Information capacity of the Hopfield model[J]. IEEE Transactions on Information Theory, 1985, 31(4): 461-464.

[25]杨淳. 基于BP神经网络的AUV航迹预测方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2014.

[26]马国兵, 张楠. 一种基于神经网络的机动目标轨迹预测方法[J]. 青岛理工大学学报, 2006(05):108-111.

[27]杨彬, 贺正洪. 一种GRNN神经网络的高超声速飞行器轨迹预测方法[J]. 计算机应用与软件, 2015, 32(7): 239-243.

[28]Hinton G E, Osindero S, Teh Y W. A fast learning algorithm for deep belief nets[J]. Neural computation, 2006, 18(7): 1527-1554.

[29]Yilmaz I, Kaynar O. Multiple regression, ANN (RBF, MLP) and ANFIS models for prediction of swell potential of clayey soils[J]. Expert systems with applications, 2011, 38(5): 5958-5966.

[30]Geng Z Q, Qin L, Han Y M, et al. Energy saving and prediction modeling of petrochemical industries: A novel ELM based on FAHP[J]. Energy, 2017, 122: 350-362.

[31]Wang J, Zhang C. Software reliability prediction using a deep learning model based on the RNN encoder–decoder[J]. Reliability Engineering & System Safety, 2018, 170: 73-82.

[32]Chen K, Zhou Y, Dai F. A LSTM-based method for stock returns prediction: A case study of China stock market[C]//2015 IEEE international conference on big data (big data). IEEE, 2015: 2823-2824.

[33]Fu R, Zhang Z, Li L. Using LSTM and GRU neural network methods for traffic flow prediction[C]//2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC). IEEE, 2016: 324-328.

[34]Abdulnabi A H, Wang G, Lu J, et al. Multi-task CNN model for attribute prediction[J]. IEEE Transactions on Multimedia, 2015, 17(11): 1949-1959.

[35]Valsamis A, Tserpes K, Zissis D, et al. Employing traditional machine learning algorithms for big data streams analysis: The case of object trajectory prediction[J]. Journal of Systems and Software, 2017, 127: 249-257.

[36]Min K, Kim D, Park J, et al. RNN-Based Path Prediction of Obstacle Vehicles With Deep Ensemble[J]. IEEE Transactions on Vehicular Technology, 2019, 68(10): 10252-10256.

[37]Xin L, Wang P, Chan C Y, et al. Intention-aware long horizon trajectory prediction of surrounding vehicles using dual lstm networks[C]//2018 21st International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2018: 1441-1446.

[38]Altché F, de La Fortelle A. An LSTM network for highway trajectory prediction[C]//2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2017: 353-359.

[39]Mao S, Tu E, Zhang G, et al. An automatic identification system (AIS) database for maritime trajectory prediction and data mining[M]. Proceedings of ELM-2016. Springer, Cham, 2018: 241-257.

[40]石庆研, 岳聚财, 韩萍, 等. 基于LSTM-ARIMA模型的短期航班飞行轨迹预测[J]. 信号处理, 2019(12).

[41]张凯. 基于海量出租车轨迹数据的学习与预测[D]. 西安: 长安大学, 2019.

[42]Hsu P M, Zhu Z W. Car trajectory prediction in image processing and control manners[C]//2016 IEEE International Conference on Intelligent Transportation Engineering (ICITE). IEEE, 2016: 45-49.

[43]Fu X, Jiang Y, Lu G, et al. Probabilistic trajectory prediction in intelligent driving[J]. IFAC Proceedings Volumes, 2014, 47(3): 2664-2672.

[44]Feng X, Cen Z, Hu J, et al. Vehicle trajectory prediction using intention-based conditional variational autoencoder[C]//2019 IEEE Intelligent Transportation Systems Conference (ITSC). IEEE, 2019: 3514-3519.

[45]Wang Z, Nie Z, Sheng G. Dynamic Position Predicting of Underactuated Surface Vessel with Unscented Kalman Filter[C]//2018 Chinese Automation Congress (CAC). IEEE, 4030-4033.

[46]邱洪生. 基于卡尔曼滤波的船舶航行轨迹异常行为预测算法研究[D]. 天津: 河北工业大学, 2012.

[47]Zhang X, Liu G, Hu C, et al. Wavelet Analysis Based Hidden Markov Model for Large Ship Trajectory Prediction[C]//2019 Chinese Control Conference (CCC). IEEE, 2019: 2913-2918.

[48]王垒, 宋庭新. 内河航道船舶避碰轨迹规划与预测[J]. 湖北工业大学学报, 2019, 34(159): 67-71.

[49]陈麒龙, 邵哲平. 基于非等时距时序残差ARIMA模型的船舶行为预测[C]//2019世界交通运输大会.

[50]徐婷婷, 柳晓鸣, 杨鑫. 基于BP神经网络的船舶航迹实时预测[J]. 大连海事大学学报, 2012(01): 11-13.

[51]任宇翔, 赵建森, 刘卫, 等. 基于AIS数据和LSTM网络的船舶航行动态预测[J]. 上海海事大学学报, 2019, 40(03): 32-37.

[52]谢新连, 陈紫薇, 魏照坤, 等. 基于极限学习机的船舶航行行为预测[J]. 重庆交通大学学报(自然科学版), 2019, 38(08): 7-12+123.

[53]何正伟, 杨帆, 刘力荣. 基于AIS数据的船舶安全航行水深参考图[J]. 交通运输工程学报, 2018, 18(04): 175-185.

[54]关克平, 江靖楠, 吴天裕, 等. 基于OSG和Bullet的船舶碰撞响应三维仿真[J]. 上海海事大学学报, 2014, 35(4): 1-6.

[55]中华人民共和国国家质量监督检验检疫总局/中国国家标准化管理委员会. GB/T 20068-2017. 船载自动识别系统(AIS)技术要求[S]. 2017.

[56]Ester M, Kriegel H P, Sander J, et al. Density-based spatial clustering of applications with noise[C]//Int. Conf. Knowledge Discovery and Data Mining. 1996, 240: 6.

[57]McKinley S, Levine M. Cubic spline interpolation[J]. College of the Redwoods, 1998, 45(1): 1049-1060.

[58]刘立群, 吴超仲, 褚端峰, 等. 基于Vondrak滤波和三次样条插值的船舶轨迹修复研究[J]. 交通信息与安全, 2015, 000(004): 100-105.

[59]Huang Y, Li Y, Zhang Z, et al. GPU-accelerated compression and visualization of large-scale vessel trajectories in maritime IoT industries[J], IEEE Internet of Things Journal, 2020, 7(11): 10794-10812.

[60]Schuster M, Paliwal K K. Bidirectional recurrent neural networks[J]. IEEE transactions on Signal Processing, 1997, 45(11): 2673-2681.

[61]Graves A, Schmidhuber J. Framewise phoneme classification with bidirectional LSTM and other neural network architectures[J]. Neural Networks, 2005, 18(5): 602-610.

[62]Graves A, Liwicki M, Bunke H, et al. Unconstrained on-line handwriting recognition with recurrent neural networks[C]//In J. Platt, D. Koller, Y. Singer, and S. Roeeis, editors, NIPS’2007, pages 577-584.

[63]Graves A, Schmidhuber J. Offline handwriting recognition with multidimensional recurrent neural networks[C]//In D. Koller, D. Schuurmans, Y. Bengio, and L. Botton, editors, NIPS’2008, pages 545-552.

[64]Graves A, Mohamed A, Hinton G. Speech recognition with deep recurrent neural networks[C]//In ICASSP’ 2013, pages 6645-6649.

[65]Baldi P, Brunak S, Frasconi P, et al. Exploiting the past and the future in protein secondary structure prediction[J]. Bioinformatics, 1999, 15(11): 937-946.

[66]Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.

[67]Cho K, Van Merri?nboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. arXiv preprint arXiv:1406.1078, 2014.

[68]Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The journal of machine learning research, 2014, 15(1): 1929-1958.

[69]Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[J]. arXiv preprint arXiv:1502.03167, 2015.

[70]Kingma D P, Ba J. Adam: A method for stochastic optimization[J]. arXiv preprint arXiv:1412.6980, 2014.

[71]Helbing D, Molnar P. Social force model for pedestrian dynamics[J]. Physical review E, 1995, 51(5): 4282.

[72]郝美薇, 戴华林, 郝琨. 基于密度的K-means算法在轨迹数据聚类中的优化[J]. 计算机应用, 2017, 37(010): 2946-2951.

[73]江玉玲, 熊振南, 唐基宏. 基于轨迹段DBSCAN的船舶轨迹聚类算法[J]. 中国航海, 2019(3): 1-5.

[74]朱军. 船舶交通管理基础[M]. 大连: 大连海事大学出版社, 2012.

[75]Fujii Y, Tanaka K. Traffic capacity[J]. The Journal of navigation, 1971, 24(4): 543-552.

[76]Goodwin E M. A statistical study of ship domains[J]. The Journal of navigation, 1975, 28(3): 328-344.

[77]Davis P V, Dove M J, Stockel C T. A computer simulation of marine traffic using domains and arenas[J]. The journal of Navigation, 1980, 33(2): 215-222.

[78]Davis P V, Dove M J, Stockel C T. A computer simulation of multi-ship encounters[J]. The Journal of navigation, 1982, 35(2): 347-352.

[79]Coldwell T G. Marine traffic behaviour in restricted waters[J]. The Journal of Navigation, 1983, 36(3): 430-444.

[80]Zhao J S, Wu Z L, Wang F C, et al. Comments on Ship Domains[J]. Journal of Navigation, 1993, 46(3): 422-436.

[81]Kijima K, Furukawa Y. Design of automatic collision avoidance system using fuzzy inference[J]. IFAC Proceedings Volumes, 2001, 34(7): 65-70.

[82]Smierzchalski R. On-Line Trajectory Planning in Collision Situations at Sea by Evolutionary Computation-Experiments[J]. IFAC Proceedings Volumes, 2001, 34(7): 407-412.

[83]Pietrzykowski Z, Uriasz J. The ship domain-a criterion of navigational safety assessment in an open sea area[J]. The Journal of Navigation, 2009, 62(1): 93.

[84]Wilson P A, Harris C J, Hong X. A line of sight counteraction navigation algorithm for ship encounter collision avoidance[J]. The Journal of Navigation, 2003, 56(1): 111.

[85]Kao S L, Lee K T, Chang K Y. A fuzzy logic method for collision avoidance in vessel traffic service[J]. The Journal of Navigation, 2007, 60(1): 17.

[86]?mierzchalski R, Michalewicz Z. Modeling of a ship trajectory in collision situations at sea by evolutionary algorithm[J]. IEEE Transaction on Evolutionary Computation, 2000, 4(3): 227-241.

[87]Szlapczynski R, Szlapczynska J. Review of ship safety domains: Models and applications[J]. Ocean Engineering, 2017, 145: 277-289.

[88]Wang N. An intelligent spatial collision risk based on the quaternion ship domain[J]. The Journal of Navigation, 2010, 63(4): 733.

中图分类号:

 U675    

条码号:

 002000034377    

馆藏号:

 YD10010313    

馆藏位置:

 203    

备注:

 403-西院分馆博硕论文库;203-余家头分馆博硕论文库    

无标题文档

   建议浏览器: 谷歌 火狐 360请用极速模式,双核浏览器请用极速模式