Dynamic Event-Triggered-Based Optimal Performance Constraint Control for Autonomous Vehicles via Reinforcement Learning

This paper investigates the optimal tracking control problem for autonomous vehicles with performance constraints and external disturbances. To begin with, an asymmetric barrier Lyapunov function is introduced to address the position and velocity constraints of autonomous vehicles. Terminal time constraints, combined with prescribed performance functions, ensure the rapid and accurate achievement of optimal control objectives. Then, the Sub-Actor-Critic framework is established by using optimal backstepping, which yields the solution of the Hamilton-Jacobi-Bellman equation. Subsequently, the identifier neural networks are utilized to estimate external disturbance. Additionally, a dynamic event-triggered mechanism is integrated to reduce trigger frequency by intermittently relaying error signals, which helps to minimize actuator stress and communication load. Finally, through Lyapunov stability analysis, it is confirmed that the proposed tracking control strategy ensures all signals within the closed-loop system are uniformly ultimately bounded, and effectively preventing the occurrence of the Zeno phenomenon. Numerical simulation verify the effectiveness and superiority of this strategy.