This paper deals with the development of a robust nonlinear control strategy for direct-driven Permanent Magnet Synchronous Generator (PMSG) wind turbines. First-Order Sliding Mode Controllers (FOSMC) are commonly used in such systems to regulate generator rotational speed and optimize power extraction; however, they encounter significant drawbacks, notably the undesirable chattering effect and the necessity for measuring additional system states, leading to increased complexity and implementation costs. In addressing these issues, an effective control technique based on an Adaptive Super-Twisting Algorithm (ASTA) has been proposed. This algorithm not only mitigates the chattering effect but also offers a more cost-effective implementation by incorporating online gains adaptation to accommodate the unpredictable nature of wind speeds. The proposed control technique has been evaluated in MATLAB/Simulink through model-in-the-loop simulations; demonstrating superior performance compared to classical FOSMC and STA controllers under various scenarios.