Neuro-fuzzy compliance control for rehabilitation robotics

This paper presents a methodology for the design of a real-time neuro-fuzzy controller for the robotic rehabilitation of patients with upper-limb dysfunction due to neurological diseases. The approach utilizes a fuzzy-logic schema to introduce compliance into the human-robot interaction, and to allow the emulation of a wide variety of therapy techniques. It also allows for the fine-tuning of system dynamics using linguistic variables. The rule base for the system is trained using a fuzzy clustering algorithm and applied to experimental data gathered during traditional therapy sessions. The compliance rule base is combined with a hybrid neuro-fuzzy compensator to automatically tune the dynamics of the system on-line. The control algorithm is implemented as a platform-independent solution to facilitate the rehabilitation of patients using multiple manipulator configurations. Preliminary experimentation has shown promising results indicating that the proposed methodology can accurately replicate the desired compliance profiles in real-time.