An intelligent robot control system for physiotherapic applications has been developed. The intelligent robot control system consists of a specially designed robotic hand with built-in sensors, an interfacing module between the robot system and the computer, an intelligent path planning module and a fuzzy logic based intelligent control module. The robotic hand with the integrated palm and two fingers has been used to perform the padding and kneading opeartions. The sensory information of the robotic hand have been used in the intelligent control process. The intelligent path planning and control modules have been constructed with the knowledge bases (KBS) and the fuzzy logic based inference mechanism, which are able to deal with uncertainties by manipulating the fuzzy terms. Thus, with the fuzzy/linguistic input terms, the required parameters can be generated for the path planning module. The massaging path can be planned by using the KBS in the intelligent path planning module. While the task execution is monitored by the intelligent control module. The intelligent control module allows error-correction strategies to be formulated. The required corrections can be carried out by using the on-line KBS and fuzzy inference mechanism in the intelligent control module. Experimental results are presented, which show the feasibility and the effectiveness of the designed intelligent control system.