Exoskeleton robots are wearable electromechanical structures which can work interacting with human limbs. Theserobotsareused as assistivelimbs, rehabilitationandpoweraugmentationpurposesforelderly or paralyzedpersons andhealthypersons respectively.Human body neuro-muscular system varies the stiffness and damping of the human joints regularly and thus provides flexible and stable movement capability with minimum energy consumption. Damping coefficients and torques in the joints needs to be adjusted with the change of stiffness to provide stable behavior during walking cycles. Magneto-rheological brake are the passive actuators that can adjust the damping torques in a very short response time. However, the maximum values of damping coefficients and torques are needed for the design of MR dampers to be used in the joints of exoskeleton robots. Shamaei et. al. (2013) derived a set of statistical equations to predict thejoint stiffnessin a gait cycle for the persons with different height and weights. In this paper, the maximum values of damping coefficients and torques are calculated for the design of MR dampers to provide stability in the joints of exoskeleton robots. The results can be used as initial design criteria of MR dampers which will be added to the biomimetic joints of exoskeleton robots, prostheses, ortheses and humanoid robots