An Adaptive Proportional Navigation Algorithm based on BADS under Solar Illumination Constraint

This paper proposes an adaptive guidance algorithm based on Bayesian Adaptive Direct Search (BADS) to address the limitations of fixed-parameter proportional navigation guidance laws in multi-target tracking scenarios under solar illumination constraints in geostationary Earth orbits (GEO). First, under illumination constraints, a generalized proportional navigation law regulates the relative line-of-sight (LOS) angle rate dynamically between the spacecraft and targets. Second, the BADS optimization mechanism autonomously adjusts the guidance coefficients by integrating hybrid Bayesian inference with mesh-adaptive direct search, ensuring efficient parameter tuning. Third, a dual-phase strategy, integrating global exploration with local grid polling, improves convergence efficiency and ensures adaptability to dynamically changing mission conditions. Simulations demonstrate a miss distance of 0.3 m (84.8 % lower than conventional methods) and a convergence time of 2.5 seconds, outperforming particle swarm optimization (PSO) by 31.7 %. These results confirm the algorithm's enhanced precision and adaptability for complex spaceborne missions.