ADAPTIVE ASSISTIVE ADMITTANCE CONTROL FOR ROBOTIC UPPER LIMB REHABILITATION

Mohammad Farajtabar, Marie Charbonneau

Keywords: Human-Robot Interaction, Adaptive Admittance Control, Rehabilitation

Abstract: Advances in robotics and AI are making robots integral to daily life, underscoring the importance of studying Human-Robot Interaction. In healthcare, particularly rehabilitation, robots can serve as effective and reliable assistants. We propose an assistive variable admittance control method for upper-limb rehabilitation, in which users guide the robot’s end-effector along a desired path by physically grasping it. The controller adapts the admittance parameters based on the magnitude and direction of the interaction force, thereby supporting more precise motion. In addition, an assistance mode activates when the interaction force deviates from the desired direction toward the target or when the end-effector strays from the reference trajectory. Passivity analysis is employed to ensure stable human-robot interaction. Experimental results show that the proposed method reduces trajectory tracking error and enhances rehabilitation performance with minimal robot interference. The robot provides assistance only when necessary, while during the rest of the exercise it acts as a passive follower, allowing users to rely on their own strength.

Submission Number: 43