APPLICATION OF ARTIFICIAL INTELLIGENCE FOR DEFORMATION PREDICTION IN GEODETIC MONITORING
Keywords:
artificial intelligence; machine learning; deep learning; deformation monitoring; geodetic monitoring; GNSS; InSAR; remote sensing; structural health monitoring; landslide early warning.Abstract
Geodetic monitoring of engineering structures and natural slopes increasingly relies on continuous, multi-sensor data streams produced by GNSS reference networks, satellite radar interferometry, and terrestrial laser scanning. The volume and heterogeneity of these data exceed the practical capacity of classical deterministic and statistical deformation models, motivating the adoption of artificial intelligence (AI) and machine learning (ML) techniques. This paper reviews how AI methods — ranging from gradient boosting and support vector regression to recurrent neural networks, attention mechanisms, and ensemble stacking — are applied to forecast deformation in dams, bridges, buildings, open-pit mines, and landslide-prone terrain. Particular attention is given to hybrid architectures that combine signal decomposition with deep learning, and to the integration of GNSS time series with InSAR-derived displacement fields. The discussion highlights demonstrated gains in prediction accuracy and early-warning lead time, alongside persistent constraints related to model interpretability, transferability between monitoring sites, and the scarcity of labeled failure events. The paper concludes by outlining prospects for physics-informed networks, multimodal sensor fusion, and pretrained foundation models for deformation time series, arguing that AI is becoming a structural component of modern geodetic monitoring rather than an auxiliary analytical tool
References
[1] Dong, X., Ma, R., Tian, W., Zhao, K., Zhong, Z., Zhu, J., Sun, G., & Xiang, H. (2026). High-precision prediction method for mine deformation based on GNSS RTK and stacking ensemble learning. Scientific Reports, 16, 12277. https://doi.org/10.1038/s41598-026-41945-x
[2] Gao, W., Li, Z., Chen, Q., Jiang, W., & Feng, Y. (2022). Modelling and prediction of GNSS time series using GBDT, LSTM and SVM machine learning approaches. Journal of Geodesy, 96(10), 71. https://doi.org/10.1007/s00190-022-01662-5
[3] Strnad, D., Mongus, D., Horvat, Š., & Šegina, E. (2025). A multi-task deep learning approach for landslide displacement prediction with applications in early warning systems. Scientific Reports, 15. https://doi.org/10.1038/s41598-025-29084-1
[4] Xie, Y., Meng, X., Wang, J., Li, H., Lu, X., Ding, J., Jia, Y., & Yang, Y. (2024). Enhancing GNSS deformation monitoring forecasting with a combined VMD-CNN-LSTM deep learning model. Remote Sensing, 16(10), 1767. https://doi.org/10.3390/rs16101767
[5] Zhang, R., Gao, C., Pan, S., & Shang, R. (2020). Fusion of GNSS and speedometer based on VMD and its application in bridge deformation monitoring. Sensors, 20(3), 694. https://doi.org/10.3390/s20030694
[6] Zhou, C., Ye, M., Xia, Z., Wang, W., Luo, C., & Muller, J.-P. (2025). An interpretable attention-based deep learning method for landslide prediction based on multi-temporal InSAR time series: A case study of Xinpu landslide in the TGRA. Remote Sensing of Environment, 318, 114580.
