Abstract—This paper proposes a reinforcement learning (RL)-based event-triggered robust optimal control method for mobile Euler-Lagrange systems with both dead-zone and saturation from actuators. Firstly, kinematics and dynamics of the system are integrated into the equivalent system, where both of the dead-zone and saturation inputs are treated. Secondly, event-triggered robust optimal control and dead-zone disturbance laws are designed, where their parameters are only updated when a triggering condition occurs. Via RL techniques, the new triggering condition is introduced. The method not only guarantees the stability of the closed system and the convergence of the cost function to the bounded L 2 -gain optimal value but also relaxes identification procedures for unknown nonlinear functions. Additionally, it maintains the minimum inter-event time between two sequent triggering instants greater than zero, thus the Zeno’s behavior is avoided. Finally, the simulation of a nonholonomic wheeled mobile robot system with dead-zone and saturated torques is implemented to verify the effectiveness of the proposed method.