Measuring the Dynamic Impedance of the Human Arm

摘要

This thesis presents techniques to measure the human arm's mechanical impedance using a rehabilitation robot, drawing applications to telerehabilitation and patient motor function assessment. Conventional teleoperation system stability analysis and control assumes the human operator remains passive and injects no net energy into the system. The validity of this assumption is evaluated by mathematically analyzing the passivity of empirical arm impedance models. The results show that the arm is passive when relaxed but may become active when participants rigidly hold their arm in place. This non-passive behaviour originates from the central nervous system's position control response. A novel, cost-effective impedance measurement technique is also presented, in which an expensive commercial force sensor is replaced by a virtual sensor incorporating a model of the robot's dynamics and kinematics. The technique is validated on a mass-spring system of known impedance and applied to the human arm.