Abstract: ‍ ‍Interferometric synthetic aperture radar （InSAR） is a principal method for terrain height measurement， facilitating the acquisition of elevation data for ground targets and enabling the comprehensive three-dimensional perception of intricate terrain and topography. Owing to its flexible baseline characteristics， the distributed InSAR system can dynamically change the baseline configuration， breaking through the baseline limitations of single-station multi-channel InSAR. This adaptability caters to diverse terrains， enhancing the accuracy of elevation inversion and taking a prominent role in geographical mapping， resource survey， reconnaissance， and early warning. Presently， InSAR systems predominantly rely on spaceborne formations， limiting configuration significantly. Several studies have focused on airborne systems； however， the existing methods are predominantly based on two stations with a single baseline configuration， which exhibits low inversion accuracy， particularly for complex and steep terrains. In response to this shortcoming， we propose a distributed multi-baseline UAV-borne InSAR elevation inversion system based on hybrid baselines， along with improvements in InSAR technology to adapt to the proposed novel architecture. This study establishes the geometric configuration of the distributed multi-baseline UAV InSAR system and derives the principle underlying the elevation inversion of the system. Subsequently， the technical advantages of this system in phase unwrapping under flexible configuration scenarios are compared with those of existing systems， and its implementation method is derived. Theoretical arguments are presented to demonstrate the suitability of the proposed flexible configuration interferometric system for complex and steep scenarios. Finally， through simulation experiments conducted on both simulated terrain and real mountainous terrain， this study attained elevation inversion of ground object scenes using the distributed multi-baseline UAV-borne InSAR system， thereby validating the feasibility of the technology. Additionally， this study leveraged a comparative analysis with existing systems to highlight the superiority of the proposed system in terrain elevation information inversion for complex terrains.