The interest in unmanned excavator is growing more and more for the sake of work efficiency and safety of the operator. It is general understanding that an optimal working path planning and inverse dynamic control should be resolved first for the realization of the unmanned excavator. In this paper, the methods to determine the optimal working path based on minimum torque or time and to simulate digging works tracking on the designed working path are proposed. In case of the minimum torque, the optimal working path is determined to minimize the joint torques of attachments. On the other hand, in case of minimum time, the optimal working path is decided to minimize time duration for one-cycle working considering the hydraulic constraints such as oil flow rate and relief pressure etc. In order to verify the inverse dynamic code used in the optimization and the optimized path, field measurements were carried out for the various parameters such as swing angle, cylinder lengths, and pressures of a swing motor and hydraulic cylinders during digging works. The modified FEE(Fundamental Earthmoving Equation) is adopted for the modeling of soil-tool interaction in digging process and the inverse dynamics with external force such as constant and reducing lifting weights is used in lifting and dumping process. Finally, all the simulated data were compared with the measured data in order to examine the effectiveness of the proposed methods in view of the development of the unmanned excavator.