As ocean resources continue to be increasingly exploited, autonomous underwater vehicles (AUVs) are finding expanding applications. However, the complex and dynamic nature of underwater environments presents challenges. Addressing the trajectory tracking and disturbance rejection issues of AUVs in two-dimensional underwater environments, a disturbance rejection model predictive control (DRMPC) algorithm is proposed. This algorithm combines model predictive control (MPC) with a sliding mode observer (SMO) to form a composite controller, enabling AUVs to effectively overcome unknown disturbances and accomplish trajectory tracking while satisfying coupling constraints. Through comparative simulation experiments, it is demonstrated that the proposed DRMPC can more effectively achieve trajectory tracking control for two-dimensional AUVs.