Fatigue loads on wind turbines caused by an asymmetric wind field become an increasing concern when the scale of wind turbines increases. This thesis presents a preview-based approach to reduce asymmetric loads by using Light Detection And Ranging (LIDAR) measurements. A Model Predictive Controller (MPC) is developed that utilize a preview history of the wind obtained from the LIDAR. The controller is based on a model with individual pitching of the blades, such that the asymmetric loads can be reduced by cyclic blade pitching. Using a transformation of moments acting on the blades, the controller is able to determine the pitching of the blades. This is done while still maintaining a given power reference and satisfying a set of actuator constraints. The designed controller was tested on a 5 MW wind turbine in the FAST simulation software and compared to the same controller without LIDAR data and a baseline controller for the turbine. The results showed that MPC without LIDAR performed similarly to the baseline controller and that MPC with LIDAR was able to reduce the asymmetric loads while still maintaining the power reference.