UAV Formation Optimal Synchronous Control Based on Uncertainties Nonlinear Dynamics

Aiming at the second-order UAV multi-agent system with unknown nonlinear dynamics, this paper introduces an enhanced reinforcement learning optimal control method based on the actor-critic-identifier (ACI) framework to achieve the position-velocity consistency synergy of leader-follower. To address the algorithmic redundancy issue caused by the complex Hamilton–Jacobi–Bellman (HJB) equation solving in traditional reinforcement learning between the critic and actor networks, the ACI framework incorporates identifiers which simplifies the design of adaptive laws and eliminates the need for prior knowledge of system dynamics and continuous excitation conditions. Theoretical analyses demonstrate the Lyapunov-stable and asymptotically convergent error in consistency for the closed-loop system.Simulations validate the approach’s effectiveness.