Model free based finite time fault‐tolerant control of robot manipulators subject to disturbances and input saturation

Abstract

AbstractThis article proposed a model free based finite time fault‐tolerant control (FTC) method of robot manipulators subject to unknown disturbances, input saturation, and actuator faults. First, an accuracy‐driven integral terminal sliding mode surface (ADITSMS) was constructed, and the finite‐time convergence and the parametric analysis were investigated. Second, based on the proposed ADITSMS, a double‐loop control framework which consists of the position loop and velocity loop was proposed, and the finite time command filtered algorithm was utilized to estimate the derivative of the virtual control law of the position loop, thus a novel finite‐time robust adaptive FTC was proposed. Third, to eliminate the negative effects of the faults and disturbances and implement model free control, the uncertainty and disturbance estimation is designed, and a novel adaptive algorithm is developed to suppress the estimation error whose upper bound or derivative was not required. Meanwhile, an auxiliary system is constructed to deal with the input saturation. The proposed controller possess the superiorities of model free nature and global finite time convergence property. Finally, numerical simulations and comparison studies were conducted to demonstrate the effectiveness of the proposed controller.