This paper presents a disturbance observer-based control (DOBC) framework for nonlinear systems subject to disturbances. The proposed approach is divided into two stages: first, a nonlinear controller is designed to stabilize the nonlinear plant and achieve performance specifications such as tracking or regulation. Second, a nonlinear disturbance observer is designed to estimate and compensate for external disturbances. The paper establishes global exponential stability for the nonlinear disturbance observer under certain conditions and provides a semiglobal stability result for the composite controller. The effectiveness of the nonlinear DOBC approach is demonstrated through the control of a two-link robotic manipulator, where the performance is significantly improved compared to a computed torque controller without disturbance attenuation. The theoretical results include the global exponential stability of the nonlinear disturbance observer and the semiglobal exponential stability of the composite controller.This paper presents a disturbance observer-based control (DOBC) framework for nonlinear systems subject to disturbances. The proposed approach is divided into two stages: first, a nonlinear controller is designed to stabilize the nonlinear plant and achieve performance specifications such as tracking or regulation. Second, a nonlinear disturbance observer is designed to estimate and compensate for external disturbances. The paper establishes global exponential stability for the nonlinear disturbance observer under certain conditions and provides a semiglobal stability result for the composite controller. The effectiveness of the nonlinear DOBC approach is demonstrated through the control of a two-link robotic manipulator, where the performance is significantly improved compared to a computed torque controller without disturbance attenuation. The theoretical results include the global exponential stability of the nonlinear disturbance observer and the semiglobal exponential stability of the composite controller.