Hybrid multi-objective control allocation strategy for reusable launch vehicle in re-entry phase

Abstract

The reentry of the reusable launch vehicle (RLV) normally requires redundant and heterogeneous actuators to assist attitude control, so the control allocation (CA) technology for the over-actuated system is a key research nowadays. This paper not only discusses how to allocate the control command between two different kinds of actuators, the reaction control system (RCS) and control surfaces, but also presents a hybrid multi-objective control allocation strategy based on the analytic hierarchy process (AHP) method to achieve the pareto solutions to RLV's allocation problems. Firstly, both actuators' dynamic models with corresponding constraints, and RLV's nonlinear model are built as the basis. Secondly, this paper considers a series of optimization functions, such as the minimum of control allocation error, the minimum of the fuel consumption, the maximum of actuators' control efficiency, which makes RLV's allocation problem multi-objective and multi-constraint. Meanwhile, to achieve the pareto solution or the integrated optimal one from all the feasible solutions, a weighted combination approach is introduced in. Then, this multi-objective allocation problem is transformed into a more easily solved, single-objective one, in which each optimization function is multiplied with its corresponding dynamic weight value based on the preference matrix of AHP. Finally, this designed strategy is demonstrated in RLV's re-entry phase, which not only verifies its effectiveness and fault-tolerant capability, but also shows its advantages of achieving better comprehensive allocation performance by comparisons with the single-objective control or the other two traditional daisy-chaining and proportional allocation methods.