A Multi-Source Data Fusion Framework for Low-Altitude Turbulence Intensity Modeling and Its Application in Route Planning

Abstract

With the rapid expansion of the low-altitude economy, accurately monitoring low-altitude turbulence, a significant threat to flight safety, has become a critical challenge. Existing meteorological instruments, such as wind profiler radars, Doppler weather radars, and automatic surface stations, cannot measure turbulence directly. Moreover, their data exhibit significant heterogeneity in spatiotemporal resolution and physical parameters, making it difficult to generate high-precision, high-resolution 3D turbulence intensity distributions. To address this issue, this study proposes a novel multi-source data fusion framework. This framework first constructs unified turbulence proxy metrics for different devices (surface stations, wind profilers, weather radars). It then employs a spatiotemporal interpolation algorithm based on geographical information to unify the heterogeneous data onto a regular 3D grid. Finally, a comprehensive turbulence intensity index () is generated through an adaptive weighted fusion strategy. We applied this fusion model to a case study in the Nanjing area of China, producing high-resolution 3D turbulence products with a horizontal resolution of 100m and a vertical resolution of 50m. Experimental results demonstrate that the fusion method can reveal the spatial structure of turbulence more comprehensively compared to using any single data source. As a downstream application, we successfully integrated the fused turbulence field into a route planning system where turbulence avoidance is the primary objective, utilizing the A* algorithm to plan safe and optimal paths for low-altitude aircraft. This research provides a reliable and practical technical solution for the refined monitoring and early warning of low-altitude turbulence, holding significant value for ensuring the safety of low-altitude operations.