This paper presents a hybrid control strategy for robotic leg prosthesis. We first synthesise an artificial gait based on human data. A mathematical dynamic model of robotic leg having three degree of freedom, i.e., one each at hip, knee and ankle joint is developed here. Since the dynamics of leg is a highly complex and nonlinear system, radial basis neural network is trained offline for inverse dynamics of leg and is used as a part of control strategy. The simulation environment contains a model of the robotic leg dynamics, a subsystem for artificial gait synthesis, three independent neural network and PD controllers to control the movements of three joints separately in order to achieve the level walking. The simulation work is carried out in Matlab 6.5 and Simulink 5.0. The results showed that the gait of the prosthetic leg can be controlled to a near normal one for level walking.