A bio-inspired optimization decentralized control mechanism for on-orbit target service by implicit self-reconfiguration of space modular self-reconfigurable satellites

The traditional on-orbit service of the space modular self-reconfigurable satellite (SMSRS) is constrained by the necessity for established space environment information and the predetermined fixed target configuration, resulting in reduced flexibility and efficiency. The maintenance and capture missions of on-orbit targets by implicit self-reconfiguration (ISR) of SMSRS in uncertain, complicated environments lacking definitive final configurations present a novel yet exceptionally complex and challenging problem. To this end, we propose a bio-inspired optimization decentralized control mechanism (BIODCM) for space modular self-reconfigurable satellites in space. This mechanism integrates novel gradient optimization models and motion control optimization strategies. The gradient optimization models involve a target gradient optimization model that provides local predictions for guiding directional reconfiguration of uncertain topological configurations towards the target position using heuristic information from the top and root, a peel gradient model that determines the motion permission of modules by constructing hierarchical relationships between modules through local information exchange and gradient directions, and a fractal gradient model for wrapping the target. Depending on the specially designed CA rules, the motion control optimization strategies include an extension motion control strategy that directs the movable module to the current optimal position with the highest comprehensive target gradient, and a fractal grasp motion control strategy that helps the movable module select the nearest predicted fractal position to occupy for covering the target surface. The intelligent swarm of SMSRS, adaptable to different configuration sizes, can auto-establish an environment-responsive configuration and achieve target approach and grasping. The implicit self-reconfiguration process utilizing the BIODCM method presented in this paper exhibits a superior global workspace (approaching 100% of the maximum theoretical Manhattan extension distance), an accelerated convergence rate, and a reduced number of moving modules and transfer steps in comparison to the current distributed control mechanism employing L-systems and Cellular Automata (LSCA) and the rapid self-reconfiguration motion planning optimization (RSRMPO) method. Self-reconfiguration simulation experiments with random on-orbit targets demonstrate that the BIODCM method is not only more adaptable to complicated environments but also facilitates implicit self-reconfiguration more swiftly and with reduced energy consumption, thereby establishing significant advantages for intelligent orbital servicing and deep-space exploration missions.