The robot has to adapt its movement with the various condition of surfaces and ensuring the stability with the proper motion of its bust and legs in order to be able to move on different surfaces. A lot of basic parameters of the gait can be expressed by the planned seven-linked bipedal robot. One of the issues that have always attracted the attention of researchers, in this field, is to predict the motion path that guarantees the stability and minimizes the energy. In this study, parameter optimization being used which means that at the first, joint angles are defined as a parameter functions and then with respect to kinematics constraints that define maximum and minimum of joint angles, the problem of motion obtained in the way that maximizing stability of robot and minimizing energy and puts the robot in the permitted stable region. Also, we tried to have zero moment point scale used to calculate the stability of robot. Experimental tests in order to motion analysis for walking healthy were performed and the results were validated. Finally, due to the presented model and predicted path, the robot can move like a person. Comparison of experimental results and the result of presented model were used in each step to validate the accuracy of the proposed method.