Bipedal robots due to similar physical and motion structure with a human body and the applications that can be imagined for them due to this similarity have been focused by many researchers and research teams around the world. In the construction and implementation of bipedal robots, different fields of engineering including mechanical design, electronics in order to set up different communication, processing and visual parts and control to maintain stability and balance are used. Bipedal robot control can be divided into two parts. The first part is aimed at optimum tracking for each joint using a suitable controller and the second part is to modify the initial trajectory to maintain the stability of the robot. The first part is called low-level control and the second part is known as high-level control. In this thesis, first the bipedal robot is discussed with dynamic equations and then the computed torque and fuzzy-supervisory controllers are designed to select the best performance in the presence of disturbances. The considered disturbance is a non-modeled bump on the path of the robot so that the robot will be able to pass it without modifying trajectory and the performance of the controllers is compared in passing this bump. Then the trajectory modifying methods based on the workspace using force sensors and fuzzy systems is developed. by using this method the robot will be able to pass the bumps with unknown height or the depth. Also, the direct adaptive fuzzy method is applied to remove unwanted vibrations of the robot’s upper body angle using the hip joint of support leg and online ideal trajectory modification. Keyword: trajectory modification, force sensors, direct adaptive fuzzy method, disturbance