Microgrids (MGs) typically function in two operational modes: on-grid and off-grid. The primary challenges in the off-grid mode involve voltage and frequency fluctuations or instability. To address these issues, an appropriate control system is required to regulate the microgrid’s voltage and frequency. Considering the system’s nonlinear characteristics, parameter uncertainties, and load variations, the use of nonlinear and robust control techniques is deemed an effective strategy. In this study, a nonlinear adaptive super-twisting controller based on sliding mode control (SMC) theory is developed to manage the microgrid. The main motivation for employing this controller is its ability to mitigate distortions under parameter uncertainties. To this end, the dynamic equations of the system are first derived, and subsequently, a controller is designed using the principles of SMC to ensure overall system control. To overcome the chattering phenomenon, an adaptive super-twisting algorithm is proposed. The closed-loop stability of the system is verified through the Lyapunov stability theorem. The effectiveness of the proposed method is assessed using MATLAB/Simulink simulations, and its performance is compared with conventional control approaches. The results demonstrate that the proposed adaptive super-twisting controller offers superior performance and practical applicability compared to traditional methods. The proposed controller is evaluated in an islanded (off-grid) DC microgrid configuration under load steps, irradiance variations, and battery parameter uncertainties. Simulation results demonstrate superior voltage regulation compared to a conventional super-twisting sliding mode controller and a PI controller, achieving a settling time below 0.18 s, an overshoot below 0.4%, and an RMSE of 0.42 V.