The problem of wind turbine control has received significant attentions in recent years. The maximization of energy generation and load reduction are two major objectives for the control of wind turbine systems. Varieties of approaches have been proposed to achieve the above two objectives. Specifically, indirect speed control (ISC) and PID control have been widely used for the energy maximization purpose, while H infinity control has been applied to reduce the load of the wind turbine when the wind speed is high. However, the above approaches did not fully consider various types of uncertainties of the system. It is well known that the wind turbine system is highly nonlinear and uncertain. For example, the Cp curve which determines the energy coefficient in terms of the pitch angle and rotor speed is not known exactly and may change with time. And the values of the dampings of the flexible blades, tower and shafts are also impossible to be obtained exactly. These uncertainties add to the difficulties and complexities of the energy maximization and load reduction controls of wind turbine systems. Traditional methods are too simplistic to deal with complicated uncertainties. This project aims at developing a holistic approach to the control of wind turbine systems with uncertainties. Three control methodologies will be seamlessly integrated. Namely, adaptive robust control (ARC) is applied as the feedback law to deal with parametric uncertainties and disturbances of the system. At the same time, extreme seeking control (ESC) is utilized to generate the optimal desired rotor speed and pitch angle online when the Cp curve is assumed to be unknown, while robust H infinity control law is synthesized for the perturbed pitch angle and electric torque input to reduce the loads suffered by the blades and shaft, respectively. Multiple objectives can be achieved simultaneously for the system with strong effects of uncertainties using the proposed holistic approach. To demonstrate the applicability of the new algorithm, a 3-bladed CART turbine will be used for simulation in FAST. Comparisons will be made between the proposed approach and the previous approaches.