Skip to Main Content
This paper addresses the controller design of dynamic walking on the spot and describes the experimental approach in a 5-DOF under-actuated planar biped robot. The angle and the angular velocity of the robot with respect to the ground were measured at a specific time, and the next landing position of a swing leg was controlled for the dynamically stable walking on the spot. The next landing position was determined by a numerical procedure, as walking is composed of a single support phase and an impact phase. The validity of the control method was shown by a simulation using a mathematical model regarding various initial conditions and via as experimental approach using an actual robot. In the experiment, the planar biped robot successfully walked dynamically on the spot for long time. Given that a real experimental system is different from a mathematical model, the controller is tuned using experimental try and error. In a later project, this control method will be extended for dynamic forward and backward walking.