Abstract: Aiming at the cooperative formation tracking problem for multi-UAVs under switching topologies, this paper proposes an optimal control method based on prescribed-time performance constraints. Firstly, a more flexible prescribed performance function is designed, and an improved prescribed-time error transformation mechanism is introduced to ensure that the required tracking accuracy is achieved within the prescribed time. Secondly, an identification-critic-action three-network architecture based on neural network approximation is utilized to directly solve the HJB equation, enabling adaptive online updates of the control strategy and thereby obtaining the approximate optimal control law. The stability of the closed-loop error system is analyzed by constructing multiple Lyapunov functions. Finally, simulations verify that the proposed formation tracking control method ensures that the tracking error converges to the prescribed precision within the prescribed time.