Adaptive Variable Admittance Control of Manipulator for Detumbling a Space Instable Satellite

It is a challenge for manipulator to quickly achieve detumbling of a space instable satellite (SIS) with low joint torque. In this paper, an adaptive variable admittance control method is proposed to address the above issue. Firstly, the on-ground experimental process of capturing and detumbling the SIS is introduced. Then, steady state error of the built Cartesian pose admittance controller is analyzed. An adaptive variable admittance controller is designed and the necessary and sufficient conditions of its stability and convergence are proved. After that, multi-objective optimization function for detumbling is established and non-dominated sorting genetic algorithm II (NSGA-II) based on general regression neural network (GRNN) is developed to optimize the parameters of proposed controller. Finally, numerical simulations and the ground detumbling experiment are conducted to validate the proposed method. Simulation results show that the proposed control method can effectively reduce detumbling torque and shorten detumbling time and the controller optimized by GRNN and NSGA-II significantly improves detumbling capacity. The comparison with other adaptive compliance control methods shows that the proposed method has better performance, with the 9.42% reduction in detumbling torque and the 13.25% reduction in detumbling time. Ground experiment data shows that the manipulator operates stably adopting the adaptive variable admittance control method and can effectively reduce the angular velocity of the spin SIS.