AI-ENABLED MULTI-SENSOR FUSION FOR FIELD DETECTION OF CHEMICAL WARFARE AGENTS: A SYSTEM-ORIENTED EXPERIMENTAL STUDY
DOI:
https://doi.org/10.68302/std2026.vol1.145Keywords:
CWA detection, data fusion, machine learning, multi-sensor systemsAbstract
The paper presents a system-level framework for AI-enabled multi-sensor fusion in chemical warfare agent (CWA) detection, combining ion mobility spectrometry (IMS), portable Raman spectroscopy, and electrochemical sensing within a unified machine learning pipeline. The main contribution is twofold: (i) the formulation of a modular and deployment-oriented fusion architecture that explicitly incorporates environmental variability, and (ii) a fully reproducible Python-based experimental workflow enabling direct comparison between unimodal and fused sensing strategies. A synthetic dataset, constructed from literature-informed response characteristics of representative CWA simulants and benign interferents, is used to emulate realistic operational conditions, including overlapping class distributions and environmental perturbations. The proposed feature-level fusion approach integrates heterogeneous sensor outputs into a unified feature space and applies multinomial logistic regression for probabilistic classification. Experimental results demonstrate that the fusion model significantly outperforms individual sensing modalities, achieving 99.2% accuracy and a weighted F1-score of 0.992, compared to 94.7% (Raman), 92.8% (electrochemical), and 81.4% (IMS). The analysis is further supported by uncertainty estimation, confidence interval approximation, and ablation-based evaluation, confirming the complementary contribution of each sensing modality and the robustness of the fused representation. While the current validation is based on synthetic data, the proposed framework provides a transparent, scalable, and computationally efficient foundation for future laboratory validation with simulants and real-world deployment in CBRN scenarios. The work bridges the gap between individual sensor studies and integrated operational systems, with explicit consideration of safety, reproducibility, and regulatory constraints.
Supporting Agencies
This scientific report is funded by the Ministry of Education and Science in implementation of the Scientific Program „Security and Defence“, adopted by Decree No. 731 of 21.10.2021 and pursuant to Agreement No. D01-74/19.05.2022. The authors acknowledge the support and resources provided by Defence Institute for facilitating this study and appreciate the insightful discussions and feedback.Downloads
References
[1] OPCW, "Handbook on Chemicals," Organisation for the Prohibition of Chemical Weapons, 2024.
[2] A. Ahrens, N. Reinecke, S. Zimmermann, and A. Wieczorek, "Detection of chemical warfare agents with a miniaturized high-performance drift tube ion mobility spectrometer using high-energetic photons for ionization," Analytical Chemistry, vol. 94, no. 43, pp. 14688-14695, 2022, doi: 10.1021/acs.analchem.2c03422.
[3] T. Satoh, R. Sekioka, Y. Seto, H. Matsumiya, and H. Nakagawa, "Ion mobility spectrometric analysis of vaporous chemical warfare agents by the instrument with corona discharge ionization ammonia dopant ambient temperature operation," Analytica Chimica Acta, vol. 865, pp. 39-52, 2015, doi: 10.1016/j.aca.2015.02.004
[4] T. Kondo, R. Hashimoto, Y. Ohrui, R. Sekioka, T. Nagomi, F. Muta, and Y. Seto, "Analysis of chemical warfare agents by portable Raman spectrometer with both 785 nm and 1064 nm excitation," Forensic Science International, vol. 291, pp. 23-38, 2018, doi: 10.1016/j.forsciint.2018.07.032.
[5] A. Hakonen, P. O. Andersson, M. S. Schmidt, T. Rindzevicius, and M. Kall, "Explosive and chemical threat detection by surface-enhanced Raman scattering: A review," Analytica Chimica Acta, vol. 893, pp. 1-13, 2015, doi: 10.1016/j.aca.2015.04.010.
[6] J. Disley, G. Gil-Ramírez, J. Gonzalez-Rodriguez, A review of sensing technologies for nerve agents, through the use of agent mimics in the gas phase: Future needs, TrAC Trends in Analytical Chemistry, Volume 168, 2023, 117282, ISSN 0165-9936, https://doi.org/10.1016/j.trac.2023.117282.
[7] V. Kumar, H. Kim, B. Pandey, T.D. James, J. Yoon, and E. V. Anslyn, "Chemical warfare agent sensors and simulants: a legacy of the 21st century," Chemical Society Reviews, vol. 52, pp. 3152-3207, 2023, doi: 10.1039/D2CS00651K.
[8] M. Xu, X. Hu, H. Zhang, T. Miao, L. Ma, J. Liang, Y. Zhu, H. Zhu, Z. Cheng, and X. Sun, "Detection and pattern recognition of chemical warfare agents by MOS-based MEMS gas sensor array," Sensors, vol. 25, no. 8, 2633, 2025, doi: 10.3390/s25082633.
[9] C. M. Bishop, Pattern Recognition and Machine Learning. New York, NY, USA: Springer, 2006.
[10] S. Haykin, "Cognitive radio: brain-empowered wireless communications," IEEE Journal on Selected Areas in Communications, vol. 23, no. 2, pp. 201-220, 2005, doi: 10.1109/JSAC.2004.839380.
[11] D. L. Donoho, "High-dimensional data analysis: The curses and blessings of dimensionality," AMS Math Challenges Lecture, 2000.
[12] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed., Springer, 2009.
[13] A. Gelman et al., Bayesian Data Analysis, 3rd ed., CRC Press, 2013.
[14] Yuan, Z., Luo, X. and Meng, F. (2024), Machine Learning-Assisted Research and Development of Chemiresistive Gas Sensors. Adv. Eng. Mater., 26: 2400782. https://doi.org/10.1002/adem.202400782R.
[15] Gutierrez-Osuna, "Pattern analysis for machine olfaction: A review," IEEE Sensors Journal, vol. 2, no. 3, 2002.
[16] Aixue Li, Haoyu Yang, Wenxin Yu, Tianyang Liu, Bin Luo, Chunjiang Zhao, Application of machine learning to improve the accuracy of electrochemical sensors: A review, TrAC Trends in Analytical Chemistry, Volume 193, 2025, 118469, ISSN 0165-9936, https://doi.org/10.1016/j.trac.2025.118469.
[17] Zachary J. Baum, Xiang Yu, Philippe Y. Ayala, Yanan Zhao, Steven P. Watkins, and Qiongqiong Zhou, Artificial Intelligence in Chemistry: Current Trends and Future Directions, Journal of Chemical Information and Modeling 2021 61 (7), 3197-3212, DOI: 10.1021/acs.jcim.1c00619S.
[18] Marco and A. Gutierrez-Galvez, "Signal and data processing for machine olfaction and chemical sensing," IEEE Sensors Journal, vol. 12, no. 11, pp. 3189-3214, 2012, https://doi.org/10.1109/JSEN.2012.2192920.
[19] Yanchen Li, Zike Wang, Tianning Zhao, Hua Li, Jingkun Jiang, Jianhuai Ye, Electronic nose for the detection and discrimination of volatile organic compounds: Application, challenges, and perspectives, TrAC Trends in Analytical Chemistry, Volume 180, 2024, 117958, ISSN 0165-9936, https://doi.org/10.1016/j.trac.2024.117958.
[20] Soo-Yeon Cho, Youhan Lee, Sangwon Lee, Hohyung Kang, Jaehoon Kim, Junghoon Choi, Jin Ryu, Heeeun Joo, Hee-Tae Jung, and Jihan Kim, Finding Hidden Signals in Chemical Sensors Using Deep Learning, Analytical Chemistry 2020 92 (9), 6529-6537, DOI: 10.1021/acs.analchem.0c00137
[21] Pannone, A., Raj, A., Ravichandran, H. et al. Robust chemical analysis with graphene chemosensors and machine learning. Nature 634, 572–578 (2024). https://doi.org/10.1038/s41586-024-08003-w
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Iliyan Hutov, Ivan Ivanov, Hristo Hristov
This work is licensed under a Creative Commons Attribution 4.0 International License.
