People who have lost their ability to walk, shake hands or even talk due to brain stroke may revert their ability back to normal or near normal with the assistance of rehabilitation robots. Literatures in the field of the rehabilitation studies shown that, in most cases, repeated movements of the patient’s member can help restore the function of the injured member in patients, in which, the use of robots can be very effective. This paper aims to investigate the ability of a wearable robot for rehabilitation of injured fingers. The method of motion in human fingers hand joints for physical therapy purposes has been investigated and being considered as the optimal route of the design process and for the robot movement. This robot has been designed with the right choice of kinematic loops, where the robot pursues the optimal path to rehabilitation, while does not comes in direct contact with the patient, provides the tolerant for the patient to feel free and comfortable. In addition to the above-mentioned advantages, the designed robot reduces the number of operators as well as the number of parameters necessary, to change and control the algorithm, for the different patients. After representing the robot’s conceptual design, the kinematics analysis of the robot was dealt with. In the next step, the dynamic equations of the robot were extracted. Ultimately, using the control methods, the position, stability and efficiency of the rehabilitation robot was simulated and constructed along the desired pathway.