«Previous Article|Table of Contents|Next Article»

Journal of Marine Science and Application[ISSN:1002-2848/CN:61-1400/f]

卷:

期数:

2026年1

页码:

228-238

栏目:

出版日期:

2026-02-25

Info

Title:

Extreme Attitude Prediction of Amphibious Vehicles Based on Improved Transformer Model and Extreme Loss Function

Author(s):

Qinghuai Zhang¹;  Boru Jia¹;  Zhengdao Zhu¹;  Jianhua Xiang¹;  Yue Liu²;  Mengwei Li²

Affilations:

Author(s):

Qinghuai Zhang¹;  Boru Jia¹;  Zhengdao Zhu¹;  Jianhua Xiang¹;  Yue Liu²;  Mengwei Li²

1. School of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081, China;
2. China North Vehicle Research Institute, Beijing 100072, China

Keywords:

Amphibious vehicle; Attitude prediction; Extreme value loss function; Enhanced transformer architecture; External information embedding

分类号:

-

DOI:

10.1007/s11804-025-00705-5

Abstract:

Amphibious vehicles are more prone to attitude instability compared to ships, making it crucial to develop effective methods for monitoring instability risks. However, large inclination events, which can lead to instability, occur frequently in both experimental and operational data. This infrequency causes events to be overlooked by existing prediction models, which lack the precision to accurately predict inclination attitudes in amphibious vehicles. To address this gap in predicting attitudes near extreme inclination points, this study introduces a novel loss function, termed generalized extreme value loss. Subsequently, a deep learning model for improved waterborne attitude prediction, termed iInformer, was developed using a Transformer-based approach. During the embedding phase, a text prototype is created based on the vehicle’s operation log data is constructed to help the model better understand the vehicle’s operating environment. Data segmentation techniques are used to highlight local data variation features. Furthermore, to mitigate issues related to poor convergence and slow training speeds caused by the extreme value loss function, a teacher forcing mechanism is integrated into the model, enhancing its convergence capabilities. Experimental results validate the effectiveness of the proposed method, demonstrating its ability to handle data imbalance challenges. Specifically, the model achieves over a 60% improvement in root mean square error under extreme value conditions, with significant improvements observed across additional metrics.

References:

[1] Arjovsky M, Bottou L (2017) Towards principled methods for training generative adversarial networks. Machine Learning, ArXiv abs/1701.04862
[2] Bengio S, Vinyals O, Jaitly N, Shazeer N (2015) Scheduled sampling for sequence prediction with recurrent neural networks. Proceedings of the 29th Annual Conference on Neural Information Processing Systems (NIPS), Montreal, Canada
[3] Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research 16: 321-357. https://doi.org/10.1613/jair.953
[4] Chen Q, Zheng S, Li M, Zhuo H (2021) Research on prediction error of ship rolling motion. Ship Engineering 43(2): 42-47
[5] Ding DZ, Zhang M, Pan XD, Yang M, He XN, Assoc Comp M (2019) Modeling extreme events in time series prediction. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, USA, 1114-1122. https://doi.org/10.1145/3292500.3330896
[6] Fan JQ, Fan YY, Barut E (2014) Adaptive robust variable selection. Annals of Statistics 42(1): 324-351. DOI: 10.1214/13-AOS1191
[7] Haan L, Ferreira A (2006) Extreme value theory: An introduction. Springer, New York, USA
[8] Harrou F, Dairi A, Dorbane A, Sun Y (2024a) Enhancing wind power prediction with self-attentive variational autoencoders: A comparative study. Results in Engineering 23: 13. https://doi.org/10.1016/j.rineng.2024.102504
[9] Harrou F, Zeroual A, Kadri F, Sun Y (2024b) Enhancing road traffic flow prediction with improved deep learning using wavelet transforms. Results in Engineering 23: 14. https://doi.org/10.1016/j.rineng.2024.102342
[10] Hittawe MM, Harrou F, Togou MA, Sun Y, Knio O (2024) Time-series weather prediction in the Red sea using ensemble transformers. Applied Soft Computing 164: 16. https://doi.org/10.1016/j.asoc.2024.111926
[11] Hornik K, Stinchcombe M, White H (1989) Multilayer feedforward networks are universal approximators. Neural Networks 2(5): 359-366. https://doi.org/10.1016/0893-6080(89)90020-8
[12] Hou X, Xia S (2024) Short-term prediction of ship roll motion in waves based on convolutional neural network. Journal of Marine Science and Engineering 12(1): 102. https://doi.org/10.3390/jmse12010102
[13] Liu XY, Wu J, Zhou ZH (2008) Exploratory undersampling for class-imbalance learning. IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics) 39(2): 539-550
[14] Ma X, Chang L (2015) Investigation on nonlinear rolling dynamics of amphibiousvehicle under wind and wave load. Journal of Measurement Science and Instrumentation 6(3): 275-281
[15] Nie Y, Nguyen NH, Sinthong P, Kalagnanam J (2022) A time series is worth 64 words: Long-term forecasting with transformers. Machine Learning, arXiv preprint arXiv:2211.14730
[16] Rosenblatt M (1956) Remarks on some nonparametric estimates of a density function. Ann. Math. Stat. 27(3): 832-837. DOI: 10.1214/aoms/1177728190
[17] Ross TY, Dollár G (2017) Focal loss for dense object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, USA, 2980-2988
[18] Shafiq M, Gu ZQ (2022) Deep residual learning for image recognition: A survey. Applied Sciences-Basel 12(18): 43. https://doi.org/10.3390/app12188972
[19] Silverman BW (1986) Density estimation for statistics and data analysis. J. Roy. Stat. Soc. Ser. C 37: 120-121
[20] Sun Y, Wong AK, Kamel MS (2009) Classification of imbalanced data: A review. International Journal of Pattern Recognition and Artificial Intelligence 23(4): 687-719
[21] Tashkova K, Silc J, Atanasova N, Dzeroski S (2012) Parameter estimation in a nonlinear dynamic model of an aquatic ecosystem with meta-heuristic optimization. Ecological Modelling 226: 36-61. https://doi.org/10.1016/j.ecolmodel.2011.11.029
[22] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser L, Polosukhin I (2017) Attention is all you need. Proceedings of the 31st Annual Conference on Neural Information Processing Systems (NIPS), Long Beach, USA
[23] Wang G, Han B, Sun W (2017) Short-term prediction of ship motion based on LSTM. Ship Science and Technology 39(13): 69-72. (in Chinese)
[24] Xu G, Zhou J, Yao X, Guo Q (2005) Toss motion of amphibious vehicle in sea-way. Acta Armamentarii 26(4): 433-437. (in Chinese)
[25] Xue YF, Liu YJ, Ji C, Xue G (2020) Hydrodynamic parameter identification for ship manoeuvring mathematical models using a Bayesian approach. Ocean Engineering 195: 106612. https://doi.org/10.1016/j.oceaneng.2019.106612
[26] Yin JC, Perakis AN, Wang N (2018) A real-time ship roll motion prediction using wavelet transform and variable RBF network. Ocean Engineering 160: 10-19. https://doi.org/10.1016/j.oceaneng.2018.04.058
[27] Zhang M, Ding DZ, Pan XD, Yang M (2023) Enhancing time series predictors with generalized extreme value loss. IEEE Transactions on Knowledge and Data Engineering 35(2): 1473-1487. DOI: 10.1109/TKDE.2021.3108831
[28] Zhou HY, Zhang SH, Peng JQ, Zhang S, Li JX, Xiong H, Zhang WC (2021) Informer: Beyond efficient transformer for long sequence time-series forecasting. Machine Learning, arXiv: 2012.07436
[29] Zhou T, Ma ZQ, Wen QS, Wang X, Sun L, Jin R (2022) FEDformer: Frequency enhanced decomposed transformer for long-term series forecasting. Proceedings of the 39th International Conference on Machine Learning (ICML), Baltimore, USA

Memo

Memo:

Received date:2024-10-20;Accepted date:2025-4-1。<br>Foundation item:Supported by the National Defense Basic Scientific Research Program of China.<br>Corresponding author:Jianhua Xiang,Email:E-mail:xiangjh@bit.edu.cn

Commonly used function

Last Update: 2026-03-09