Utilizing artificial intelligence for National Transportation Safety Board unmanned aerial vehicle accident analysis and categorization

Abstract

The rapid increase in unmanned aerial vehicle (UAV) usage has introduced significant safety challenges, including issues such as system failure, loss of control, transmission failures, and collisions. Analyzing these incidents has been challenging due to the absence of a dedicated category field in the National Transportation Safety Board (NTSB) data. This research tackles this problem by utilizing artificial intelligence (AI) to automate the classification of UAV accident reports collected between 2006 and 2023. Using natural language processing techniques, we categorize NTSB reports to improve the analysis and interpretation of incident data. We also employ advanced data visualization tools to reveal geographic and temporal patterns, offering a detailed view of UAV accident trends. The results indicate that system and component failures unrelated to propulsion systems (system/component failure or malfunction [non-powerplant]) and abnormal contact upon landing (abnormal runway contact) are predicted as the primary categories (37%) of UAV accidents for the period. These insights suggest the potential value of AI-driven categorization and visualization techniques in enhancing UAV safety standards and supporting policy development. Initial results provide promising insight into the use of language models for text classification in aviation safety problems.

Keywords

UAV accident analysis / AI categorization / GPT-4 analysis / Data visualization in safety / NTSB accident data / Accident trend analysis

Acknowledgments

None.

Funding

None.

Conflict of interest

The authors declare that they have no competing interests.

Author contributions

Conceptualization: Eugene Pik

Formal analysis: Joao S. D. Garcia

Investigation: Eugene Pik

Methodology: Eugene Pik

Writing - original draft: Eugene Pik

Writing - review & editing: Joao S. D. Garcia

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data

1. The original data presented in this study are openly accessible at https://www.ntsb.gov/Pages/AviationQueryV2.aspx.

2. The data analysis scripts for this paper are available at https://doi.org/10.5281/zenodo.10576209.

Further disclosure

Part of findings has been presented in a conference “ERAU Discovery Days” (https://commons.erau.edu/discovery-day/db-discovery-day-2024/poster-session-2/54/).

References

[1] Pik E. Commercial and Recreational Active UAV Fleet in the U.S. 2016-2022; 2024. doi: 10.5281/zenodo.10573914

[2] Ferrigan J. Safety Risk Assessment for UAV Operation; Available from: https://www.regulations.gov/document/faa-2022-0426-0004 [Last accessed on 2024 Jan 15].

[3] Zhang X, Srinivasan P, Mahadevan S. Sequential deep learning from NTSB reports for aviation safety prognosis. Saf Sci. 2021;142:105390. doi: 10.1016/j.ssci.2021.105390

[4] Yang C, Huang C. Natural language processing (NLP) in aviation safety: Systematic review of research and outlook into the future. Aerospace. 2023; 10(7):600. doi: 10.3390/aerospace10070600

[5] Kasprzyk PJ, Konert A. Reporting and investigation of unmanned aircraft systems (UAS) accidents and serious incidents. Regulatory perspective. J Intell Robot Syst. 2021; 103(1):3. doi: 10.1007/s10846-021-01447-6

[6] Nguyen C, Sagan V, Bhadra S, Moose S. UAV multisensory data fusion and multi-task deep learning for high-throughput maize phenotyping. Sensors (Basel). 2023; 23(4):1827. doi: 10.3390/s23041827

[7] Nanyonga A, Wild G. Impact of Dataset Size and Data Source on Aviation Safety Incident Prediction Models with Natural Language Processing. In: 2023 Global Conference on Information Technologies and Communications (GCITC); 2023. p. 1-7. doi: 10.1109/GCITC60406.2023.10426284

[8] Lázaro FL, Madeira T, Melicio R, Valério D, Santos LFFM. Identifying human factors in aviation accidents with natural language processing and machine learning models. Aerospace. 2025; 12(2):106. doi: 10.3390/aerospace12020106

[9] New MD, Wallace RJ. Classifying aviation safety reports: Using supervised natural language processing (NLP) in an Applied Context. Safety. 2025; 11(1):7. doi: 10.3390/safety11010007

[10] Sarkar NI, Gul S. Artificial Intelligence-based autonomous UAV networks: A survey. Drones. 2023; 7(5):322. doi: 10.3390/drones7050322

[11] Kovanič Ľ, Topitzer B, Peťovský P, Blišťan P, Gergeľová MB, Blišťanová M. Review of photogrammetric and lidar applications of UAV. Appl Sci. 2023; 13(11):6732. doi: 10.3390/app13116732

[12] Filipe, Anema F, Story R, et al. Python-Visualization/Folium: V0.15.1; 2023. doi: 10.5281/zenodo.10255171

[13] Caswell TA, Droettboom M, Lee A, et al. Matplotlib REL: V3.5.1; 2021. doi: 10.5281/zenodo.5773480

[14] Waskom M. Seaborn: Statistical Data Visualization; 2021. doi: 10.5281/zenodo.4645478

[15] National Transportation Safety Board. NTSB Aviation Investigation Search; 2023. Available from: https://www.ntsb.gov/pages/aviationqueryv2.aspx [Last accessed on 2024 Jan 11].

[16] Pilgrim M. Chardet: Universal Encoding Detector for Python 3; 2023. Available from: https://pypi.org/project/chardet [Last accessed on 2024 Jan 15].

[17] Solc T. Unidecode: ASCII Transliterations of Unicode Text; 2024. Available from: https://pypi.org/project/unidecode [Last accessed on 2024 Jan 15].

[18] Van Rossum G, Drake FL. The Python Language Reference; Available from: https://docs.python.org/3/reference/index.html [Last accessed on 2024 Jan 15].

[19] Harris CR, Millman KJ, Van der Walt SJ, et al. Array programming with NumPy. Nature. 2020; 585(7825):357-362. doi: 10.1038/s41586-020-2649-2

[20] The Pandas Development Team. Pandas-dev/Pandas: Pandas; 2023. doi: 10.5281/zenodo.3509134

[21] OpenAI. OpenAI Platform-Documentation; 2024. Available from: https://platform.openai.com/docs/api-reference/introduction [Last accessed on 2024 Jan 11].

[22] ICAO. Aviation Occurrence Categories-Definitions and Usage Notes; Available from: https://www.ntsb.gov/safety/data/documents/datafiles/occurrencecategorydefinitions.pdf [Last accessed on 2024 Jan 11].

[23] Pik E. GPT-4 Assisted Categorization and Visualization of NTSB UAV Accident Reports-Python Scripts; 2024. doi: 10.5281/zenodo.10576209