Variable gain gradient descent-based reinforcement learning for robust optimal tracking control of uncertain nonlinear system with input constraints

In recent times, a variety of reinforcement learning (RL) algorithms have been proposed for optimal tracking problem continuous time nonlinear systems with input constraints. Most these are based on the notion uniform ultimate boundedness (UUB) stability, in which normally higher rates avoided order to restrict oscillations state error smaller values. However, this comes at cost convergence critic neural network weights. This paper addresses that by proposing novel tuning law containing variable gain gradient descent can adjust rate Hamilton–Jacobi–Bellman (HJB) approximation error. By allowing high could improve Simultaneously, it also results tighter residual set, trajectories augmented system converge to, leading A bound UUB stability update mechanism is proved. Numerical studies then furnished validate descent-based presented system.