Article Details

Abstract

In modern maritime battlefields, the complex and changeable environment poses numerous challenges to the accuracy of target threat assessment. To address the effective evaluation of maritime targets and the analysis of complex assessment factors, a method for maritime target threat assessment based on Deep Learning is proposed. According to the two perspectives of environmental factors and target attributes in complex maritime battlefields, an index system for maritime target threat assessment is constructed. The data types are categorized into categorical and numerical features for the analysis of maritime target threat assessment. Based on the introduced TabTransformer model, Linear Discriminant Analysis is incorporated to optimize the model, and a target threat assessment model is established. Through training and testing with simulated data, the accuracy of the model is approximately 91%. The model has strong robustness, provides a new solution for maritime target threat assessment, and has broad application prospects.

Keywords

threat assessment; Linear Discriminant Analysis; Deep Learning; Multi-head Self-attention Mechanism

References

1. [1] XIA J Y，CHEN M Q，FANG W G，et al. Normal Wiggly Probabilistic Hesitant Fuzzy Set and Its Application in Battlefield Threat Assessment [J].International Journal of Fuzzy Systems，2023，25（1）：145-167.
2. [2] ZHAO R J，YANG F B，JI L N，et al. Dynamic Air Target Threat Assessment Based on Interval-Valued Intuitionistic Fuzzy Sets, Game Theory, and Evidential Reasoning Methodology [J/OL].Mathematical Problems in Engineering，2021（1）：6652706[2024-08-12].https://doi.org/10.1155/2021/6652706.
3. [3] Sun Yunke, Fang Zhigeng, Chen Ding Multi time threat assessment based on dynamic grey principal component analysis [J]. Systems Engineering and Electronic Technology, 2021, 43 (3): 740-746
4. [4] LUO R N，HUANG S C，ZHAO Y，et al. Threat Assessment Method of Low Altitude Slow Small (LSS) Targets Based on Information Entropy and AHP [J/OL].Entropy，2021，23（10）：1292[2024-08-10].https://doi.org/10.3390/e23101292.
5. [5] YU D，WANG H J，LI B Y，PROMETHEE-Based Multi-AUV Threat Assessment Method Using Combinational Weights [J/OL].Journal of Marine Science and Engineering，2023，11（7）：1422[2024-08-06].https://doi.org/10.3390/jmse11071422.
6. [6] HUANG X，KHETAN A，CVITKOVIC M，et al. TabTransformer: Tabular Data Modeling Using Contextual Embeddings [J/OL].arXiv:2012.06678 [cs.LG].[2024-08-09].https://doi.org/10.48550/arXiv.2012.06678.
7. [7] Tang Weiguang, Xu Chao, Kang Yanxiao Maritime target threat assessment based on combinatorial weighting TOPSIS [J]. Radio Engineering, 2024, 54 (5): 1286-1293
8. [8] Zhao Ruojing, Yang Storm, and Gina Target threat assessment based on possibility theory extended grey correlation [J]. Command Information Systems and Technology, 2021, 12 (3): 23-29
9. [9] Chai Huimin, Zhang Yong, Li Xinyue, etc A deep learning based aerial target threat assessment method [J]. Journal of System Simulation, 2022, 34 (7): 1459-1467
10. [10] Jiang Dawei, Dong Yangyang, Zhang Lidong, etc Research on deep learning based aerial target threat assessment technology [J/OL]. Journal of System Simulation, 2024:1-17 [2022-06-30] https://doi.org/10.16182/j.issn1004731x.joss.23-1323.
11. [11] CAO Y，KOU Y X，XU A，et al. Target Threat Assessment in Air Combat Based on Improved Glowworm Swarm Optimization and ELM Neural Network [J/OL].International Journal of Aerospace Engineering，2021（1）：4687167[2024-09-01].https://doi.org/10.1155/2021/4687167.
12. [12] Liu Lifang, Yang Haixia, Qi Xiaogang Time frequency domain feature extraction algorithm based on linear discriminant analysis [J]. Systems Engineering and Electronic Technology, 2019, 41 (10): 2184-2190

About the authors

Wu Zhengwei (2000-), male, Han ethnicity, from Xuancheng, Anhui Province, is a master's student with a research focus on machine learning and situational awareness.