关键词: 无人机载SAR, 重复轨道差分干涉, 形变测量, 实地仿真试验

Abstract:

UAV-borne interferometric synthetic aperture radar (InSAR) systems exhibit significant advantages of high mobility and high spatial resolution in deformation monitoring along large-scale engineering projects. However, its instability in flight trajectory and attitude easily leads to difficulties in interferometric processing of repeat-pass SAR images. Based on the independently developed X-band fixed-wing vertical take-off and landing UAV-borne interferometric SAR system, this paper addresses the issues of SAR imaging and image registration that cause poor repeat-pass interferometric quality, and proposes a common heading angle fitting SAR imaging and block-wise registration method, which significantly improves interferometric coherence and deformation monitoring capability. Experimental results show that the trajectory control accuracy is better than ±2 m, enabling differential interferometry for all repeat-pass images; the common heading angle fitting SAR imaging method solves the azimuth interferometric fringe problem caused by non-parallel trajectory (spatial baselines), and the block-wise registration method increases coherence from approximately 0.6 to over 0.8. With an image resolution of 0.18×0.18 m, the deformation characteristics of two small-sized simulated settlement plates can be accurately identified, and the differences between interferometric deformation measurements and true values are less than ±2.6 mm (with root mean square errors of 1.2 mm and 1.5 mm, respectively), verifying that the proposed system and methods in this paper have engineering deformation monitoring capability with millimeter-level accuracy.

Key words: UAV-borne SAR, repeat-pass interferometric, deformation measurement, field-based simulation experiment