Adaptive optimal control for a class of uncertain systems with saturating actuators and external disturbance using integral reinforcement learning

The problem of adaptive optimal control for a class of nonlinear uncertain systems with saturating actuators and external disturbance is investigated in this paper. Considering the saturating actuators, a non-quadratic cost function is adopted. The key of this optimal control problem is to find the solution to the Hamilton Jacobi Bellman equation (HJB). An online intergral reinforcement learning (IRL) algorithm based-Neural Network (NN) is given to approximate the solution. Unlike traditional integral reinforcement learning algorithms, data onto a period of time stored together with current data are used to update the neural network weights in place of persistence of excitation (PE) condition. This method overcomes the shortcomings of the PE condition which is not easy to be checked online. Finally, numerical examples are given to show the effectiveness of the proposed methods.