Neuroadaptive backstepping tracking control of robotic manipulators considering actuator dynamics

This paper investigated two novel robust adaptive NN-based backstepping algorithms for the tracking problem of robotic rigid-link manipulators with uncertain kinematics, dynamics and actuator model. The controllers are both designed based on the backstepping scheme. Uncertainties and external disturbances in the manipulator dynamics are compensated by radial basis function in every step of backstepping. The first control algorithm could guarantee the tracking error converges to a small neighborhood of zero; the second improved control algorithm demanding link accelerations information could make sure the tracking error goes to zero quickly and precisely. The feasibility and effectiveness of the proposed methods are verified by simulation studies.