This paper presents a neural network-based adaptive sliding mode control method for unmanned surface vehicles to address the challenges of controlling nonlinear systems with uncertain dynamic parameters and the impact of unmeasured external disturbances. To address these challenges, this study employs a combination of adaptive control theory and radial basis function neural networks to approximate and compensate for system uncertainties and external disturbances. This integration facilitates the systematic synthesis of a sliding mode control law, ensuring accurate trajectory tracking for unmanned surface vehicles. This method allows the system to maintain stable control performance while significantly mitigating the chattering effect commonly associated with sliding mode control. Simulation outcomes confirm the validity and performance of the developed control strategy.