Optimal Prescribed Performance Control for Euler-Lagrange Systems

This work introduces an optimal control policy aiming to stabilize an Euler-Lagrange system of unknown dynamics, while simultaneously satisfying predefined response criteria. The proposed methodology unfolds in two stages. In the initial phase, a neural network is trained online via an iterative process to capture the system's dynamics. Subsequently, in the second phase, by utilizing the acquired dynamics, a successive approximation algorithm is employed to derive a nearly optimal control law while taking into consideration prescribed performance specifications such as convergence speed and steady state error. Validation of our method is achieved through a simulated paradigm, confirming the success of both the identification process and the minimization of the adopted integral cost.