Haptic Guidance of Light–Exoskeleton for Arm–Rehabilitation Tasks Luis I. Lugo-Villeda, Antonio Frisoli, Oscar Sandoval–Gonz´ alez, Miguel A. Padilla, Vicente Parra-Vega, Carlo A. Avizzano, Emanuele Ruffaldi and Massimo Bergamasco. Abstract— Fixed–Base Exoskeleton applications have in- creased rapidly in the last few years, evidently as part of promising rehabilitation robotic programs of the robotics worldwide community, where in particular Human–Robot– Interaction (HRI) plays an important role in its design and control because they are tightly coupled to human–limbs. Exoskeletons embrace HRI as well as technological and the- oretical challenges towards real and effective rehabilitation. In this realm, some questions arise, to name a few, what is the relationship between the exchanged energy between human and exoskeleton? How can we assess rehabilitation factors under HRI philosophy? This paper attempts to establish answers to these questions, which can be embodied into rehabilitation HRI using a Light–Exoskeleton. A compliant haptic guidance scheme for human arm subject to minimum–jerk–trajectories criterion is proposed. Preliminary experimental results provide further insight of a haptic guidance scheme taking into account decisive factors into the HRI such as human pose, haptic guidance control, reaching and tracking tasks, the complexity of the virtual environment, and muscles activity. I. I NTRODUCTION In last few decades, robot–aided–therapy has provided high performance so as to gain acceptance into the rehabilita- tion area [2] since the new trend on human–robot–interaction (HRI) furnishes more efficient tools to carry out the robot– aided–rehabilitation tasks, in comparison with manual ther- apy [10], e.g., the recovery process of a person who has suffered an stroke[7] is more effective basically because rehabilitation robotics allows a less subjective assessment in comparison when using conventional therapy approaches. Focusing in robot–aided arm therapy issues, there is a vast literature as well as commercial robots for rehabilitating the motion of human arms, wherein HRI is included in its design criteria to assist the robots, in particular within virtual environments to render more interactivity and high level of comfort during recovery sessions for the patient, [6]. This kind of rehabilitation robots for arm therapy can be classified into, a) End–Effector-based robots, and b) Exoskeletons 1 [14]. The main difference between these two approaches This manuscript was received on March 30th of 2009. This work was supported by SKILLS-IP and Scuola Superiore Sant’Anna. Luis I. Lugo-Villeda, Antonio Frisoli, Oscar Sandoval–Gonz´ alez, Miguel Padilla, Carlo A. Avizzano, Emanuele Ruffaldi and Massimo Bergamasco are with Perceptual Robotics–PERCRO , Scuola Superiore Sant’Anna, Via Martiri della Libert` a, Pisa, Italy {l.lugovilleda,a.frisoli}@sssup.it Vicente Parra-Vega is with Robotics and Advanced Manufacturing Divi- sion,Research Center for Advanced Studies Saltillo Campus. CINVESTAV Carretera Saltillo-Monterrey Km 1.5 - CP 25000 - Ramos-Arizpe, Coahuila, Mexicovparra@cinvestav.mx 1 All the cited exoskeletons belong to Upper Part Fixed–Frame Exoskele- ton (UFBEx). relies on the fact that exoskeletons are anthropomorphic robots tightly attached to human–limbs, whose joint axes fully determines the arm pose and time variations, while it is not necessarily the case for End–Effector-based robots, such as the MIT Manus[9] 2 , which trains patients who have suffered an stroke and have lost arm motor skills. In case of a), the main purpose of MIT–Manus is guiding the human– arm in virtual environments to display desired and real reaching exercise carried out by the patient. Another exam- ple of a) is the Mirror Image Movement Enabler system, [12], a PUMA 560-based robot which can impose bilateral 3–D force–position motions. Likewise, the Bi–manu-track robot for upper–limb rehabilitation uses active practice of forearm motions, such as pronation/supination, and wrist flexion/extension into mirror like fashion [17]. EU Gentle/s Project presents a 3 degree–of–freedom (DoF) haptic device for right–handed subjects, which uses virtual environments for reaching exercises to tackle the arm tracking of smooth trajectories based on polynomials, [1]. Finally, ARM-guide, [16], drives the forearm along linear position profile into 2-D space having a guided force training in joint space. Under the second classification b) we have the Armin II, an exoskeleton for human–arm rehabilitation whose 6 DoF have been used in conjunction with virtual environments, suitable for helping post–stroke patients. The exoskeleton uses a simple PD plus gravity compensation controller to obtain apparent interactive forces, which are computed in the basis of virtual environment interaction [13]. Finally, 7–DoF upper–limb exoskeleton which can be used for therapeutic diagnostics and for physiotherapy, or as a haptic device in virtual reality, is presented in [15], however it does not introduces neither the techniques for doing rehabilitation nor the control scheme for assuring stability. In the realm of HRI used for rehabilitation purposes, there arise two important facts: (i) simple control regulator schemes have been used, therefore limiting its scope and impact in demanding tasks such as rehabilitation–based robotics, an essentially a tracking task. (ii) Can we induce performance betterment when using signals coming directly from patient measurements during rehabilitation tasks? These questions are addressed this paper in the framework of hap- tic guidance using preliminary EMG patient readings with promising results. Notice that haptic guidance for UFBEx has been done indirectly with conventional controllers, however they do not exploit at maximum the exoskeleton performance since more simple controllers are implemented. In this paper 2 A pantograph-based manipulator which works into 2–D. The 18th IEEE International Symposium on Robot and Human Interactive Communication Toyama, Japan, Sept. 27-Oct. 2, 2009 ThA4.2 978-1-4244-5081-7/09/$26.00 ©2009 IEEE 903