Full-order neural sliding model control of robotic manipulator with unknown dead-zone

In this paper, a full-order neural sliding mode control (FNSMC) scheme is proposed for robotic manipulator with unknown input dead-zone. The dead-zone is transformed into a linear time-varying system to avoid the dead-zone inverse compensation. A full-order sliding mode surface is constructed to develop the finite-time controller to improve both the singularity and chattering problems. The system uncertainties including the unknown dead-zone are approximated by a radial basis function neural network (RBFNN). The finite-time stability of the closed-loop system can be guaranteed by the proposed control law via Lyapunov synthesis. Comparative simulations are provided to illustrate the effectiveness of the proposed method.