轻量化积分修正数值方法及器载计算机实现

Integral correction methods involve extensive sparse matrix computations，posing challenges for meeting onboard computing requirements under resource-constrained conditions. To address this challenge，a family of lightweight integral correction methods is proposed. Through matrix-based state vector representation，redundant sparse matrix computations are eliminated. Lightweight iterative formulations are developed to reduce matrix quantities，significantly conserving computational resources. Implementation and parallel acceleration of these methods are achieved on the FPGA module of the KZJ-02 onboard computer for solving single-spacecraft and multi-spacecraft orbital problems. It is demonstrated that the lightweight methods achieve computational efficiency improvements of at least 6. 6 and 6. 1 times in these scenarios，respectively，while maintaining higher accuracy compared to finite difference methods. The findings validate the effectiveness of the proposed approach for resource-efficient onboard orbit computation.