Current Directions in Biomedical Engineering
Joint Journal of the German Society for Biomedical Engineering in VDE and the Austrian and Swiss Societies for Biomedical Engineering
Editor-in-Chief: Dössel, Olaf
Editorial Board: Augat, Peter / Buzug, Thorsten M. / Haueisen, Jens / Jockenhoevel, Stefan / Knaup-Gregori, Petra / Kraft, Marc / Lenarz, Thomas / Leonhardt, Steffen / Malberg, Hagen / Penzel, Thomas / Plank, Gernot / Radermacher, Klaus M. / Schkommodau, Erik / Stieglitz, Thomas / Urban, Gerald A.
Controlling of a ROS-based robotic system in accordance to the assist-as-needed principle in end-effector based rehabilitation systems
Stroke is one of the most frequent diseases among the elderly and often leads to an ongoing failure of functions in the central nervous system. Due to the plasticity of the brain affected may regain lost motor function by repetitive training. Robotic devices can be an approach to accelerate the rehabilitation process by maximizing patients’ training intensity. End-effector based robotic systems are particularly suitable for this purpose and often an advantage over exoskeletons since the proximal part of the upper limb remains under the control of the patient. Furthermore, the integration of the assistas- needed principle (AAN) into these devices enables individualized, adaptable robotic support to patients during therapy. In this study an end-effector based robotic rehabilitation device based on the Robot Operating System (ROS) framework is introduced. The system allows patients to perform 3- dimensional movements without a therapist’s assistance. With regard to the AAN, focus was based on impedance control and an additional real-time adaption of the impedance control parameters by using a feedback loop. 10 healthy subjects took part in this study to evaluate the overall concept with regard to usability and quality of the supported movement. Hence, the three most promising adaption models of AAN (without adaption, adaption according to position and time, adaption according to velocity) under three different levels of movement support (0%, 50%, 100%) were investigated by administering a self-designed questionnaire and the robot kinematics. The results showed no significant differences between the three different adaption models of AAN. However, the subjective assessment of the movements was in keeping with robot kinematics and the control approaches as well as the overall system have experienced remarkable support.