A novel MPC-NNSMC composite control method for robotic manipulators considering uncertainties and constraints

Abstract

With the increasing demand for the application of robotic manipulators in complex tasks in modern industry, the requirements for their trajectory tracking control performance and robustness are becoming higher and higher. Existing robotic manipulator trajectory tracking control methods rarely comprehensively consider robust control problem caused by model parameter variations and external disturbances, and control performance optimization problem under actuator torque constraints and joint limitations. Consequently, these methods are hard to meet the control performance requirements in complex working environments and tasks. In order to solve the above limitations and problems, this paper proposes a novel composite control method. This method systematically combines model predictive control (MPC) based on a nonlinear compensation decoupling model with neural network nonsingular terminal sliding mode control (NNSMC) to simultaneously optimize the robotic manipulator trajectory tracking performance and robustness. Firstly, the radial basis function neural network is used to approximate the system uncertainty online, and then a NNSMC controller is designed on this basis to reduce the influence of model parameter perturbation and external disturbance on the robotic manipulator trajectory tracking control performance. Then, based on the system model of feedback linearization decoupling, a MPC controller is designed to meet the drive torque constraints and joint restrictions. By constructing a Lyapunov function considering the neural network estimation error and MPC constraints, the stability and feasibility of this method are analyzed. Finally, the proposed method is implemented in the robotic manipulator trajectory tracking control experiment, and the experimental results verify the effectiveness of the method. Finally, the proposed method is implemented in an actual six-degree-of-freedom robotic manipulator trajectory tracking control experiment, and the experimental results verify the effectiveness of the method.