Treffer: Fault tolerant tracking control for nonlinear systems with actuator failures through particle swarm optimization-based adaptive dynamic programming.

In this paper, an adaptive dynamic programming based fault tolerant tracking control (FTTC) method is proposed for nonlinear systems with actuator failures. To solve the tracking control problem, the considered system is augmented by combining the tracking error dynamics and the desired trajectory dynamics. By establishing a critic neural network, whose weight vector is updated by particle swarm optimization algorithm, the value function with discount factor is approximated to solve the Hamilton–Jacobi–Bellman equation, and the optimal tracking control law is derived. The neural network-based fault observer is established to compensate the control input online. Then, the augmented system states are guaranteed to be uniformly ultimately bounded under the developed FTTC method according to the Lyapunov stability theorem. Two simulation examples are provided to illustrate the effectiveness of the proposed method. • Extend the adaptive dynamic programming-based optimal tracking control for continuous-time nonlinear systems with actuator failures. • The fault tolerant tracking control strategy is achieved without constructing the improved value function. • By introducing the particle swarm optimization algorithm, the success rate of system operation is increased. [ABSTRACT FROM AUTHOR]