Hierarchical Knowledge-based Control of a Deployable Orbiting Manipulator

Abstract The advantages of combining a crisp algorithmic controller and a soft knowledge-based controller are addressed, in the specific context of controlling a space-based manipulator with flexible deployable and slewing links. A hierarchical control structure having a high-speed conventional controller at the bottom layer and an intelligent tuner at an upper layer, is developed. The top-level intelligent tuner uses a valid set of linguistic rules for adjusting proportional-derivative servos, based on fuzzy system concepts. The performance of the hierarchical control system is evaluated on the basis of the space-based simulation results. Based on the analysis it can be concluded that initial tuning of the parameters of a conventional low-level controller together with tuning during operation can improve positioning accuracy of the manipulator. The hierarchical control structure permitted desirable combination of a conventional high-bandwidth and knowledge-based low-bandwidth controllers.