In the design of control law, model-based control plays a major role for robotic manipulators, which heavily relies on an accurate parametric dynamic model. Due to the presence of the uncertainties from the robot structure and working environment, the nominal dynamic model cannot well characterize the real robot dynamics, therefore, prior to the control design, it is essential to identify the parameters of the dynamic system. This chapter presents the inertial parameter identification of the previously derived dynamic model by the least square method, in which the dynamic equation is linearized and an internal model based approach is adopted. Experimental dynamic testing is carried out to implement parameter identification and verification, for the following model-based control design.