Real-Time Onboard Geopositioning of LEO Optical Imagery via Lightweight Algorithm and FPGA- Based Heterogeneous Architecture Co-Design

Abstract

High-precision geopositioning directly determines the geometric accuracy of satellite onboard processing and product distribution. To address this requirement, this article proposes a collaborative design method for lightweight algorithms and heterogeneous architectures, enabling real-time onboard geopositioning of low-Earth orbit optical imagery on a field-programmable gate array (FPGA)-based platform. By decoupling real-time onboard geopositioning into two collaborative phases, the FPGA handles high-speed data stream interface management, auxiliary data separation, and auxiliary data capture during imaging transitions, providing a deterministic, time-synchronized data path for CPU-side computations; it performs linear prediction and the analytical solution of the rigorous geometric imaging model. Experimental validation using Gaofen-14 satellite data on the Xilinx XC5VFX130T FPGA platform demonstrates that the system achieves a real-time throughput of 5 Gbit/s. While requiring only a modest amount of FPGA logic resources, the system achieves the desired onboard horizontal and elevation accuracy. The collaborative design framework proposed in this article provides a technical roadmap for deploying physics-driven real-time geopositioning on FPGA or FPGA-based SoC platforms, offering meaningful reference value for next-generation intelligent remote sensing satellites and constellations.