This paper presents a learning controller for repetitive gait control of a biped walking robot. We propose an iterative learning control algorithm, which can learn periodic nonlinear load change which occurred according to the walking period through iterative learning, not calculating the complex dynamics of the walking robot. The learning control scheme consists of a feedforward learning rule and linear feedback control input for stabilization of the learning system. The feasibility of learning control for a biped robotic motion is shown via dynamic simulation with a 12-DOF biped walking robot.