Nonlinear Disturbance Observer Based Adaptive Neural Network Backstepping Control of Uncertain Robotic Manipulator System and Experimental Verification

This paper proposes a nonlinear disturbance observer based adaptive neural network backstepping control method for uncertain robotic manipulator system. Considering the effect of robotic manipulator system with model uncertainty on the control performances, a radial basis function neural network (RBF-NN) is introduced to estimate the model uncertainties. Meanwhile, in order to reduce the influence of estimation errors and bounded external disturbances, a nonlinear disturbance observer (NDO) is used to compensate the integrated disturbances to further improve the control performances. In the control process, in order to avoid the problem of the trivial and complex calculation and improved the computational efficiency, the filtered intermediate virtual control and the estimated compensation terms are introduced into the backstepping control. Finally, the simulation and experimental results have verified the effectiveness and superiority of the proposed method.