Fuzzy logic control of the support of a lightweight robot during rehabilitation

Abstract

The usage of robots for the rehabilitation training has the potential to relieve the therapists and to enable patients to train independently. However, the intelligent control of robots in direct interaction with humans is still subject to current research activities. This work presents an assist-as-needed control strategy for an industrial lightweight robot during upper arm rehabilitation training. To include the expert knowledge of therapists, a fuzzy logic control strategy for the stiffness of the underlying impedance controller is designed and implemented. Apart from the patient's current status, defined by pain and exhaustion, his or her general capabilities and personal characteristics are also considered by the presented approach. In order to test the designed algorithm, an experiment with an additional weight to simulate a patient suffering paresis and an oval movement in the xz-plane is done. The results illustrate that the controller is able to adapt the robot's behavior according to the current status of the patient and its general condition.