Sofia Drozd, a PhD student at the MMDA Department, presented the results of the study “Multi-Agent Explainable AI Framework for Forest Disturbance Analysis and Biomass Recovery Modeling“—conducted under the supervision of Prof. Nataliia Kussul—at the *Explainable AI in Space: Second International Workshop* (EASi 2026), held in Bremen, Germany, from August 15 to 21, 2026.
The research focuses on developing a multi-agent Explainable AI framework for analyzing forest ecosystem disturbances and modeling biomass recovery using Earth observation satellite data. The proposed approach integrates multiple AI agents to coordinate processes, build and explain models, generate code, and analyze the causes of detected forest disturbance sites.
The study results demonstrate the potential of explainable multi-agent AI systems for forest monitoring, disturbance impact analysis, and the modeling of biomass recovery scenarios.
The research is supported by international and national projects, including Horizon Europe SWIFTT and FUTUREFOR, USJRP/SNSF DT4LC, ReSeDiUm (ERDF/EFRE), and NASA, as well as projects funded by the Ministry of Education and Science of Ukraine in the fields of satellite monitoring, digital twins, environmental change modeling, and machine learning.
Full citation of the work:
Drozd, S., Kussul, N., Shelestov, A., Skakun, S. (2027). Multi-agent Explainable AI Framework for Forest Disturbance Analysis and Biomass Recovery Modeling. In: Nalepa, J., et al. Explainable AI in Space. EASi 2026. Communications in Computer and Information Science, vol. 3107. Springer, Cham. https://doi.org/10.1007/978-3-032-36806-5_5
👏 Congratulations to our colleagues on presenting their research at an international scientific forum!
