<p>This paper proposes a distributed adaptive neural network (NN) formation control strategy with an asynchronous edge-based dynamic event-triggered communication mechanism, addressing the formation control problem of networked uncertain underactuated surface vehicles (USVs) under the constraints of network communication bandwidth and the influence of external time-varying environmental disturbances. First, a fully distributed adaptive dynamic compensator with an asynchronous edge-based dynamic event-triggered communication mechanism is proposed for the distributed cooperative layer. This compensator can accurately estimate the reference signal under asynchronous intermittent communication conditions and effectively reduce the communication load. Second, in the decentralized control layer, a nonlinear dynamic disturbance compensator combining an adaptive NN is designed based on the concept of classification reconstruction. Through the mutual compensation mechanism, the effects of internal and external uncertainties on the system are effectively suppressed. Then, considering the physical response frequency limitations of actuators, an adaptive NN formation controller based on a dynamic event-triggered control strategy is proposed, which achieves precise tracking of the reference signal and alleviates the burden on the actuators. The proposed distributed formation control strategy achieves formation control in a completely distributed manner, without the need for global topology information, and strictly avoids Zeno behavior. Finally, the effectiveness of the proposed method is verified through numerical simulations and comparative analysis.</p>