Optimal Satellite Formation Reconfiguration Based on Closed-Loop Brain Storm Optimization

In recent years, satellite formation flying has become an increasingly hot topic for both the astronomy and earth science communities due to its potential merits compared with a single monolithic spacecraft system. This paper proposes a novel approach based on closed-loop brain storm optimization (CLBSO) algorithms to address the optimal formation reconfiguration of multiple satellites using two-impulse control. The optimal satellite formation reconfiguration is formulated as an optimization problem with the constraints of overall fuel cost minimization, final configuration, and collision avoidance. Three versions of CLBSOs are developed by replacing the creating operator in basic brain storm optimization (BSO) with closed-loop strategies, which facilitate search characteristic capture and enhance the optimization performance by taking advantage of feedback information in the search process. Numerical simulations are carried out using particle swarm optimization (PSO), basic BSO, and the three versions of CLBSOs. Comparison results show that all versions of CLBSOs outperform PSO and the original BSO in terms of final results and convergence speed. In addition, CLBSO reduces the computation burden and shortens CPU time to a certain extent in contrast with basic BSO. Furthermore, among the three CLBSO algorithms, the one using the strategy of difference with the best gains the best overall performance, which is inspired by the updating rule in PSO that each particle tends to move towards the individual with the best fitness.