A Therapist-Taught Robotic System for Assistance During Gait Therapy Targeting Foot Drop

Abstract

The adoption of robots in rehabilitation medicine settings has become increasingly attractive in recent years. Robots are capable of providing repetitive, high-intensity physiotherapy. In this paper, we apply kinesthetic teaching principles to a robotic system in order to allow it to first learn and then imitate a therapist's behavior while assisting a patient in a lower-limb therapy task. A therapist's assistance in lifting a patient during treadmill-based gait therapy is statistically encoded by the system using Learning-from-Demonstration (LfD) techniques. Later, the therapist's assistance is imitated by the robot, allowing the patient to continue practicing in the absence of the therapist. Preliminary experiments are performed with inexperienced users playing the role of the assisting therapist, and with healthy participants (wearing an elastic cord to simulate foot drop) playing the role of a patient. Toe clearance values are recorded, which show that the system is able to provide the full clearance needed by the patient to practice in the absence of the therapist.