Inertial motion tracking of human arm movements in stroke rehabilitation

We present a real-time monitoring system for measuring human upper limb movements in post-stroke rehabilitation. It is based on an advanced inertial sensor MT9 that is commercially available. Human upper limb motion is represented by a kinematic chain in which there are six joint variables to be considered: three for the shoulder and the others for the elbow. Kinematic models are built to study human upper limb motion in three-dimension, based on measurements of the wrist motion. A simulated annealing based optimisation method is adopted to reduce errors in measurements. Experimental results demonstrate the favorable performance of the proposed system in real motion circumstances.