Finite-time adaptive control for the dual-arm space robots with uncertain kinematics, dynamics and deadzone nonlinearities

Dual-arm space robots are capable of load transporting and coordinated manipulation for on-orbit servicing. However, achieving the accurate trajectory tracking performance is a big challenge for dual-arm robots, especially when mechanical system uncertainties exist. This paper proposes an adaptive control scheme for the dual-arm space robots with grasped targets to accurately follow trajectories while stabilizing base's attitude in the presence of dynamic uncertainties, kinematic uncertainties and deadzone nonlinearities. An approximate Jacobian matrix is utilized to compensate the kinematic uncertainties, while a radial basis function neural network (RBFNN) with feature decomposition technique is employed to approximate the unknown dynamics. Besides, a smooth deadzone inverse is introduced to reduce the effects from deadzone nonlinearities. The adaption laws for the parameters of the approximate Jacobian matrix, RBFNN and the deadzone inverse are designed with the consideration of the finite-time convergence of trajectory tracking errors as well as the parameters estimation. The stability of the control scheme is validated by a defined Lyapunov function. Several simulations were conducted, and the simulation results verified the effectiveness of the proposed control scheme.